Many kinds of subsurface defects are always present together in the subsurface of fused silica optics. It is imperfect that only one kind of defects is isolated to investigate its impact on laser damage. Therefore it is necessary to investigate the impact of subsurface defects on laser induced damage of fused silica optics with a comprehensive vision. In this work, we choose the fused silica samples manufactured by different vendors to characterize subsurface defects and measure laser induced damage. Contamination defects, subsurface damage (SSD), optical-thermal absorption and hardness of fused silica surface are characterized with time-of-flight secondary ion mass spectrometry (TOF-SIMS), fluorescence microscopy, photo-thermal common-path interferometer and fully automatic micro-hardness tester respectively. Laser induced damage threshold and damage density are measured by 351 nm nanosecond pulse laser. The correlations existing between defects and laser induced damage are analyzed. The results show that Cerium element and SSD both have a good correlation with laser-induced damage thresholds and damage density. Research results evaluate process technology of fused silica optics in China at present. Furthermore, the results can provide technique support for improving laser induced damage performance of fused silica.
©2013 Optical Society of America
The operation fluence of the transmissive optics is near the optics damage threshold for maximum output in high power laser facility such as the National Ignition Facility in the United States [1,2], the Laser MegaJoule in France  and the SGIII laser facility in China [4,5]. Every optical component can contain a number of imperfections or impurities induced by manufacture  or defects generated as a result of the environment  or excitation  conditions during its operation. These defects can cause laser damage. Literatures [9–15] show that two main kinds of defects are responsible for igniting laser damage of fused silica. One is highly absorptive contaminants (e.g., Ce, Fe, etc.) in the Beilby layer coming from polishing [9–11], the other is subsurface damage (SSD) created during grinding and/or polishing of brittle material surfaces [12–15]. These precursors decrease the laser-induced damage threshold by either high absorption of UV laser or reduction of the mechanical strength or enhancement of the local optical field . Many laboratories have developed optical defects characterization techniques to inspect defects correlated with laser damage before irradiation. The common technique to inspect contaminations is the inductively coupled plasma atomic emission spectrometer (ICP-AES) which track redeposition layer contaminants with a high sensitivity for trace elements . However, the dissolution, collection and measurement of surface impurities is more complicated to carry out accurately. There are various methods to measure SSD. Destructive measurement techniques such as polishing a taper (taper method) [16,17], a sphere (ball dimpling method)  and acid etching  has a relative high spatial resolution. Non-destructive measurement techniques such as total internal reflection microscopy (iTIRM) , white light interferometry , X ray scattering  were also studied. However the spatial resolution is too low to represent SSD actually.
There are many kinds of defects in the subsurface of fused silica optics. It is imperfect that if we only select one kind of defects to investigate its impact on laser damage neglecting the impacts of other defects. In order to understand the damage phenomena, the development of more sophisticated techniques are necessary to detect defects as extensive as possible. It is important to combine several tools which allow getting complementary information. Surface contaminations of optics component are detected by Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS)  which has a high sensitivity (on the order of magnitude of ppm to ppb for most species) for trace elements or compounds and can analyze surface and depth profiling of insulating and conducting samples. SSD is inspected by fluorescence microscopy which can be used to image defect nanostructures located in the bulk of dielectric materials for the absorption and emission characteristics [12,23]. The macroscopical characterization parameters, such as optical absorption and hardness of fused silica surface, are introduced for their effects on laser damage. Based on the characterization techniques and the measurement of laser induced damage performance, the correlation of laser damage performance with subsurface defects of fused silica are investigated.
The purpose of this work is to give a comprehensive insight into the effect of subsurface defects on laser damage of polished fused silica parts in the UV. In Section 2 of this paper, we give some information about sample manufacturing. Defects detection and characterization techniques are described in Section 3. Damage testing procedure and its results are presented in Section 4. Section 5 is devoted to the analysis and discussions of the obtained results and Section 6 draws the conclusions.
2. Sample manufacturing
Five 50mm diameter, 5mm thick Heraeus Suprasil S312 fused silica samples were manufactured by four different vendors. Cerium (Ce) oxide slurries was used in manufacturing process for all samples. However, polishing conditions were different for these samples, and so the amount and the kinds of defects are different. After polishing, one of sample was exposed to post processing to modify surface defects by use of hydrofluoric (HF) acid etching. Post processing was carried out on both sides of the sample with a constant material removal of 10 μm per side. Sample preparation methods are given in Table 1. All samples exhibited a surface roughness of ~1 nm RMS measured with a white light interferometer. Before damage testing, these samples were cleaned using same cleaning procedure.
3. Defects detection and characterization
3.1 Redeposition layer contaminants characterization: TOF-SIMS analyses
TOF-SIMS is a surface-sensitive analytical method that uses a pulsed ion beam to remove molecules from the very outermost surface of the sample. The particles are removed from atomic monolayers on the surface (secondary ions). These particles are then accelerated into a “flight tube” and their mass is determined by measuring the exact time when they reach the detector (i.e. time-of-flight). Three operational modes are available using TOF-SIMS: surface spectroscopy, surface imaging and depth profiling. In this paper we present the depth profiling of fused silica samples analyzed with an IONTOF TOFSIMS IV apparatus. By monitoring the entire mass spectrum any contamination impurities at each depth of subsurface can be detected. All sample surfaces were sputter cleaned twenty second to remove contaminations induced by surroundings.
Figure 1 shows the depth profile of Ce element detected on polished fused silica surfaces from different vendors by TOF-SIMS. The data were normalized with silicon particle number (counts 10000) as a standard. The results evaluate impurities content as a function of depth. From Fig. 1, we can see that Ce element disappears at the depth of 100 nm for all samples, and Ce content of samples is different from each other. Ce content in the subsurface layer of sample A and B is poor but that of sample C is rich. The differential is up to two orders of magnitude. Other contamination elements, such as Fe, Ca, Mg, K, Al, were monitored for maybe leading to strong absorption in the UV. The cumulated amount of each contaminant in the subsurface layer is shown in Table 2. There are rarely contamination elements in the subsurface layer of Sample A, especially without Ce, Fe and Mg. It suggests that contamination impurities introduced by polishing and manufacturing process can be removed by use of HF acid etching or other post processing.
3.2 SSD characterization: confocal fluorescence microscope analyses
Confocal fluorescence microscope is an integrated microscope system consisting of a fluorescence microscope, laser light sources, a scan head which directed the laser on the sample and collected the emission, a computer with software for controlling the scan head and display the acquisition. It allows a simultaneous measurement of both fluorescence images and bright images. Fluorescence images are detected in the 410-488nm spectral band for an excitation wavelength of 405nm and 20X objective in this paper. A 410nm high-pass filter is laid before fluorescence detector. Figure 2 shows the fluorescence image and bright image of Sample E inspected by confocal fluorescence microscope: (a) is fluorescence image and (b) is bright image. Two images are composed of 8 × 8 small images stitching. The alternating light and dark in mosaic image is due to the nonuniform intensity distribution in small images. There are many crisscross pattern cracks and point defects in fluorescence image but nothing are observed in bright image. It suggests that cracks and point defects observed in fluorescence mode are subsurface features. These features probably involve in the damage process of fused silica optics at 351 nm. SSD area of the total area percentage was introduced to characterize subsurface damage of fused silica component. SSD area of the total area percentage could be got by use of image processing and analysis in Fig. 2(a).
We also detected and analyzed the SSD of other samples. The results are shown in Table 3 which give us the SSD area percentage of fused silica samples manufactured by different vendors. The optics SSD quality is different from each other. The SSD area percentage is from 0.04% (for Sample C) to 1.99% (for Sample E).
3.3 Macroscopic characterization: photothermal absorption and hardness
We measured photothermal absorption of optics surface by photo-thermal common-path interferometer based on photothermal deflection techniques. The pump beam is CW 532 nm wavelength laser. The probe beam is a He-Ne laser. They are collinear and focused through the same objective. When pump beam pass through the sample, optical absorption induces the local temperature rise. Spatial refractive index will vary due to thermal expansion. Probe beam is deflected by the modulated refractive index gradient. The deflection of the transmitted probe beam is measured by a position sensor. Photothermal techniques has a high sensitivity for small absorption. The lowest absorptance that we can detect is about the order of magnitude of 0.01 ppm. We measured optical thermal absorption in scanning mode with the area of 1 mm2 for every sample. Intensity distribution of optical thermal absorption of Sample B is shown in Fig. 3. Because the absorption maximum position is prone to laser induced damage, the maximum absorption coefficient was selected to characterize photothermal absorption of fused silica subsurface. Table 4 shows the maximum absorption coefficient of optics components manufactured by different factory. It is clear that the maximum absorption coefficient decreases one order of magnitude for the etched Sample A. Table 4 also shows us the surface hardness of optics components manufactured by different factory. The surface hardness is measured by fully automatic micro hardness tester.
4. Damage performance measurement
A tripled Nd:YAG laser was used at a wavelength of 351 nm in our laser irradiation test equipment. The pulse is a single longitudinal mode with about 9.3 ns (FWHM), as shown in Fig. 4(a). Fluence fluctuations have a standard deviation of about ± 4.5% at 351 nm. During the test, the beam is focused on the sample surface in order to achieve high fluence. The spatial beam distribution shown in Fig. 4(b) is flat Gaussian with a diameter of 3 mm for optics damage threshold test. The modulation of irradiated area is a factor of 2.5. And the damage is always ignited at the maximum of the beam fluence. The damage thresholds tested with R on 1 are listed in Table 5. Damage was always occurred on the back surface of the samples, as verified by optical microscopic inspection.
Raster scan damage test is applied to detect the optics damage density as a function of fluence using the same laser seed. The laser is focused to provide a far field near Gaussian beam with a diameter of about 0.6 mm at 1/e2 of maximum intensity. Scan area is 10 cm2 at same fluence. Irradiated areas are detected instantaneously by a long working distance microscope in order to get the information of damage configuration. The detective sensitivity of damage size is about 10 μm. Repeating this test at several fluences on different areas permits to determine the damage density versus fluence. The damage densities of the samples versus fluence as plotted in Fig. 5 appear to follow approximate exponential growth. Figure 6 gives the damage densities of the samples at the fluence of 13.75J/cm2. Figure 5 and Fig. 6 illustrate that the growth trends of damage densities are different for the samples. The damage densities of Sample B and C increase more slowly than that of Sample D and E. Damage densities indicate defect densities that respond to laser fluence directly. The different growth trends of damage densities suggest the different type and distribution of defects.
5. Results and discussions
Subsurface defects are characterized with SSD, contaminations, optical absorption and hardness in section 3 and surface laser induced damage performance are tested in section 4. Now we analyzed and discussed the correlation existing between subsurface defects and damage performance of fused silica.
The effects of Ce, Fe, Ca, Mg, Al, K element on damage threshold and damage density are shown in Figs. 7,8,9,10,11,12, respectively. The effects of SSD, photo thermal absorption and surface hardness on damage threshold and damage density are shown in Figs. 13,14,15, respectively. In these figures, the series A, B, C, D, E are for damage threshold, and the boxed series A, B, C, D, E are for damage density. Data are fitted with an exponential decay curve for damage threshold and an exponential increase curve for damage density in Figs. 7,8,9,10,11,12,13,14,15. The data are nearer the fitted curve, which suggests that the nonlinear correlation between defect and damage is closer. Figure 7,8,9,10,11,12,13,14,15 show that Ce element and SSD both have closer relationship with damage performance. Ce element has a strong influence on damage threshold while SSD has a strong influence on damage density. Smith and Cohen reported that the value of absorption cross-section were calculated from absorption spectra reported for trace elements in Na2O•SiO2 glass . In the contaminations introduced by the grinding or polishing processes, only Ce element, Fe element and their Oxidation condition have strong absorption in UV and can cause laser damage. However, our results suggest that Fe element have a relative weak influence on damage performance, similar to Ca, Mg, Al and K element. Furthermore, the fitting curve of surface absorption and hardness are not well consonant with the test data, which suggest both surface absorption and hardness have a weak influence on damage performance.
In order to obtain the relationship between subsurface defects and damage performance, experiment data was analyzed by spearman correlation. The results are shown in Table 6. Absolute value of spearman correlation coefficient represents degree of correlation. The absolute value of 1 represents complete correlation. The absolute value of 0 represents no correlation. Negative sign represents negative correlation. Positive sign represents positive correlation. The Spearman Correlation is reliable if P is less than 0.05, otherwise the spearman correlation is not reliable. Table 6 shows us that there are complete correlations between SSD / Ce element and damage threshold / damage density. At the opposite, optical thermal absorption has a very weak correlation with damage threshold and damage density. The correlation coefficient between Fe / Ca / Al / Mg / K element and damage threshold / damage density are all −0.8 / 0.8, and hardness has correlation coefficient of 0.63 / −0.63 with damage threshold / damage density. Because P value is more than 0.05, they all have no strong influences on damage threshold and damage density. The results shown in Table 6 are consistent with the results from Figs. 7,8,9,10,11,12,13,14,15.
Compared with other samples, Sample C has the higher Ce content and the lower SSD density. Damage threshold and damage density both are quite low. These special properties would have a negative influence on the results. Therefore Sample C was not considered in Table 6. Just because of these, we can conclude that Ce element has great influence on damage threshold while SSD density has great influence on damage density. It is necessary to comprehensively analyze the correlation between subsurface defects and damage performance at the same time.
Figure 16 shows us the gray haze damage morphology induced by cerium in the interface, (a) is tested by online microscopy and (b) is tested by high power microscopy. We can see that damage sites have a diameter of ~1 um. Online microscopy with resolution of 10 um could not resolve the size of ~1 um, so damage sites look like haze. The haze damage is shallow and could not grow with subsequent laser irradiation. We neglected the damage haze when we estimated damage threshold and damage density of optics. So damage density seems to be close to SSD density. We had analyzed the correlate between cerium concentration in the interface and gray haze damage density. The gray haze damage density would decrease with cerium content.
When irradiated with sufficiently high fluence laser, defects absorbed laser energy can be heated to the critical temperature required to initiate plasma formation and ignite damage. So the optical thermal absorption should have strong correlation with damage performance. However, the results show there is very weak correlation between optical thermal absorption and damage performance. The reason maybe is relatively lower density of strong absorption defect and relatively smaller beam size of optical thermal absorption detection.
In fact it is unadvisable to analyze the correlation between one of subsurface defects and damage performance individually. SSD can serve as reservoirs for vanishingly small (femtogram) quantities of photo-absorbing impurities. SSD causes the light intensity enhancements by interference of internal reflections, which can locally reach 2 orders of magnitude . Laser induced damage would be ignited with a lower fluence if photo-absorbing impurities happen to locate at light intensity enhancements. It suggests that SSD filled with contamination is most prone to damage. The mechanical strength which can be translated with surface hardness also plays an important role in the initiation of laser induced damage. The SSD density is relative low for high-polishing fused silica optics. The test area of damage threshold is near to one square centimeter and low-density defects easily happen to meet, for example, the SSD filled with contamination. So the damage threshold is mainly decided by Ce content which is the main ignitor existed in contamination. Damage density is mainly decided by SSD for the same reason.
This paper presents the subsurface defects and 351 nm nanosecond laser damage performance of fused silica sample manufactured by different Chinese vendors. By analyzing the relationship between subsurface defects and damage performance, we conclude that SSD and Ce element in the subsurface of fused optics both have strong influence on damage threshold and damage density. Ce element has great influence on damage threshold while SSD density has great influence on damage density. Other metal impurities, hardness and optical thermal absorption all have weak correlations with damage performance. The results suggest that SSD and Ce element are responsible for igniting laser damage at present process technology of fused silica optics. It is significance for comprehensive analysis of correlation between multi-factor defects and damage performance at the same time.
The authors wish to thank Zhang Lei and Dai Chunling for contamination inspection as well as fruitful discussions, Zhang Jie and Bai Lin for subsurface defects inspection on fused silica. This work was supported by the National Natural Science Foundation of China (Grant No. 61078075) and the Science Foundation of China Academy of Engineering Physics, China (Grant No. 2011B0401065).
References and links
1. E. I. Moses, J. H. Campbell, C. J. Stolz, and C. R. Wuest, “The national ignition facility: the world’s largest optics and laser system,” Proc. SPIE 5001, 1–15 (2003). [CrossRef]
2. E. I. Moses, “National ignition facility: 1.8MJ, 750TW ultraviolet laser,” Proc. SPIE 5341, 13–24 (2004). [CrossRef]
3. M. L. André, “Status of the LMJ project,” in Solid State Lasers for Application to Inertial Confinement Fusion: Second Annual International Conference, M. L. André, ed., Proc. SPIE 3047, 38–42 (1996).
4. H. S. Peng, X. M. Zhang, X. F. Wei, W. G. Zheng, F. Jing, Z. Sui, Q. Zhao, D. Fan, Z. Q. Ling, and J. Q. Zhou, “Design of 60-kJ SG-III laser facility and related technology development,” Proc. SPIE 4424, 98–103 (2001).
5. G. Y. Xiao, D. Y. Fan, S. J. Wang, Z. Q. Lin, Y. Gu, J. Q. Zhu, Y. X. Zhen, J. Zhu, F. Q. Liu, S. C. Chen, Q. H. Chen, G. L. Huang, and X. M. Deng, “SG-II solid-state laser ICF system,” Proc. SPIE 3492, 890–895 (1999). [CrossRef]
6. S. G. Demos and M. Staggs, “Application of fluorescence microscopy for noninvasive detection of surface contamination and precursors to laser-induced damage,” Appl. Opt. 41(10), 1977–1983 (2002). [CrossRef] [PubMed]
7. S. G. Demos, A. Burnham, P. Wegner, M. Norton, L. Zeller, M. Runkel, M. R. Kozlowski, M. Staggs, and H. B. Radousky, “Surface defect generation in optical materials under high fluence laser irradiation in vacuum,” Electron. Lett. 36(6), 566–572 (2000). [CrossRef]
8. D. Ehrt, P. Ebeling, and U. Natura, “UV transmission and radiation-induced defects in phosphate and fluoride-phosphate glasses,” J. Non-Cryst. Solids 263-264, 240–250 (2000). [CrossRef]
9. J. Neauport, P. Cormont, L. Lamaignére, C. Ambard, F. Pilon, and H. Bercegol, “Concerning the impact of polishing induced contamination of fused silica optics on the laser-induced damage density at 351 nm,” Opt. Commun. 281(14), 3802–3805 (2008). [CrossRef]
10. M. R. Kozlowski, J. Carr, I. Hutcheon, R. Torres, L. Sheehan, D. Camp, and M. Yan, “Depth profiling of polishing induced contamination on fused silica surface,” Proc. SPIE 3244, 365–375 (1998). [CrossRef]
11. J. Neauport, L. Lamaignere, H. Bercegol, F. Pilon, and J. C. Birolleau, “Polishing-induced contamination of fused silica optics and laser induced damage density at 351 nm,” Opt. Express 13(25), 10163–10171 (2005). [CrossRef] [PubMed]
12. J. Neauport, P. Cormont, P. Legros, C. Ambard, and J. Destribats, “Imaging subsurface damage of grinded fused silica optics by confocal fluorescence microscopy,” Opt. Express 17(5), 3543–3554 (2009). [CrossRef] [PubMed]
13. P. E. Miller, T. I. Suratwala, L. L. Wong, M. D. Feit, J. A. Menapace, P. J. Davis, and R. A. Steele, “The Distribution of Subsurface Damage in Fused Silica,” Proc. SPIE 5991, 599101, 599101-25 (2005). [CrossRef]
14. Y. G. Li, H. Huang, R. Q. Xie, H. B. Li, Y. Deng, X. H. Chen, J. Wang, Q. Xu, W. Yang, and Y. B. Guo, “A method for evaluating subsurface damage in optical glass,” Opt. Express 18(16), 17180–17186 (2010). [CrossRef] [PubMed]
17. T. Suratwala, L. Wong, P. Miller, M. D. Feit, J. Menapace, R. Steele, P. Davis, and D. Walmer, “Subsurface mechanical damage distributions during grinding of fused silica,” J. Non-Cryst. Solids 352(52-54), 5601–5617 (2006). [CrossRef]
18. Y. Zhou, P. D. Funkenbusch, D. J. Quesnel, D. Golini, and A. Lindquist, “Effect of etching and imaging mode on the measurement of subsurface damage in microground optical glasses,” J. Am. Ceram. Soc. 77(12), 3277–3280 (1994). [CrossRef]
19. O. W. Fähnle, T. Wons, E. Koch, S. Debruyne, M. Meeder, S. M. Booij, and J. J. Braat, “iTIRM as a tool for qualifying polishing processes,” Appl. Opt. 41(19), 4036–4038 (2002). [CrossRef] [PubMed]
20. A. Wuttig, J. Steinert, A. Duparre, and H. Truckenbrodt, “Surface roughness and subsurface damage charaacterization of fused silica substrates,” in Proceedings Of The EUROPTO Conference on Topical Fabrication and Testing , Proc. SPIE 3739, 369–376 (1999). [CrossRef]
21. L. Névot and P. Croce, “Sur l’étude des couches superficielles monoatomiques par reflexion rasante (spéculaire ou diffuse) de rayons X par la méthode de l’empilement sandwich,” J. Appl. Cryst. 8(2), 304–314 (1975). [CrossRef]
22. M. A. Bolorizadeh, S. Ruffell, I. V. Mitchell, and R. Gwilliam, “Quantitative depth profiling of ultra-shallow phosphorus implants in silicon using time-of-flight secondary ion mass spectrometry and the nuclear reaction 31P(a,p0)34S,” Nucl. Instr. and Meth. in Phys. Res. B 225, 345–352 (2004).
23. S. G. Demos and M. Staggs, “Application of fluorescence microscopy for noninvasive detection of surface contamination and precursors to laser-induced damage,” Appl. Opt. 41(10), 1977–1983 (2002). [CrossRef] [PubMed]
24. H. L. Smith and A. J. Cohen, “Absorption spectra of cations in alkali-silicate glasses of high ultra-violet transmission,” Phys. Chem. Glasses 4(5), 173–187 (1963).