Increases in the laser damage threshold of fused silica have been driven by the successive elimination of near-surface damage precursors such as polishing residue, fractures, and inorganic salts. In this work, we show that trace impurities in ultrapure water used to process fused silica optics may be responsible for the formation of carbonaceous deposits. We use surrogate materials to show that organic compounds precipitated onto fused silica surfaces form discrete damage precursors. Following a standard etching process, solvent-free oxidative decomposition using oxygen plasma or high-temperature thermal treatments in air reduced the total density of damage precursors to as low as <50 cm−2. Finally, we show that inorganic compounds are more likely to cause damage when they are tightly adhered to a surface, which may explain why high-temperature thermal treatments have been historically unsuccessful at removing extrinsic damage precursors from fused silica.
© 2014 Optical Society of America
Laser damage of optics is an important limiting factor in the operation of UV lasers. In fused silica, laser damage is usually due to surface damage rather than intrinsic bulk damage [1–3 ] and occurs when a near-surface damage precursor absorbs sub-band gap light and locally heats the fused silica. The local temperature rise leads to thermally-activated runaway absorption in the surrounding silica matrix, creating a destructive absorption front that ultimately results in fracture [4, 5 ]. These fractures limit the life of fused silica optics used for high peak-power lasers, especially lasers with large apertures or contrast and modulation .
The improvement in the laser damage resistance of fused silica surfaces has been due to the progressive elimination of various classes of damage precursors. At fluences below 10 J/cm2, surface damage is caused by impurities associated with polishing (e.g., ceria) or electronic defects associated with fractures [7–9 ]. These “low-fluence” damage precursors can be effectively removed via a hydrofluoric acid etching process .
We recently reported that damage precursors at higher fluences are discrete, nanoscale, and uniformly distributed defects that have a finite surface density. Moreover, they are extrinsic to the optical surface, e.g. salts that precipitate during wet chemical processing. These damage precursors can be photoactive without being bulk light absorbers, meaning they may be composed of nearly any material. Managing and eliminating impurity sources during etching and rinsing operations reduced the density of high-fluence precursors by a factor of more than 100, from ~105 cm−2 to ~600 cm−2 .
In this work, we show that one source of remaining high-fluence precursors is trace organic contaminants found in nominally high-purity water used during wet-chemical processing. A reduction in the total damage density from 600 cm−2 to <50 cm−2 was achieved through dry oxidative decomposition of these carbonaceous impurities via O2 plasma cleaning or thermal treatments in air. Unlike mid-IR laser heating, a technique used to create locally damage-resistant fused silica surfaces, the whole-optic processes described here do not impart additional mechanical stress on the optic [12–15 ]. The identification and mitigation of this class of precursors is a significant step toward the fabrication of surface-damage resistant UV laser optics.
2. Experimental methods and materials
2.1 Sample preparation
Fused silica samples were prepared from polished 50 mm round optics (Sydor Optics or CVI-Melles Griot). As-received optics were ultrasonically cleaned first in a 10% NaOH solution for 10 minutes at 45 °C and then in a 3% detergent solution (Brulin, 1696 PNC) for 20 minutes at 55 °C, followed finally by a thorough spray rinse and air dry. Prior to etching, optics were cleaned in a 45 °C aqueous solution of 40% concentrated nitric acid and 10% hydrogen peroxide for 45 minutes, followed by a spray rinse. Samples were etched in a high-purity buffered oxide etch (BOE, Columbus Chemical Industries) to a depth of 2.5 μm, followed by immersion in an ultrasonicated water bath and rinsing in a custom-built spray chamber. Details of the process can be found elsewhere [10, 11 ].
All process water was sourced from a purification system that consists of several ion exchange beds, a UV-sterilizer, and 0.2 μm particulate filters (Pacific Water Systems). In this work, water from this system is termed “ultrapure.” The measured resistivity was >17.7 MΩ-cm. Samples were prepared in a Class 100 cleanroom, and all fixturing that contacted either process chemicals or the fused silica sample was fluoropolymer or polypropylene.
Organic contaminants were deposited onto samples by aerosolizing 40 μL of a 2 wt% water-soluble polymer solution through a pneumatic concentric nebulizer (Meinhard TR30-A3). The droplets landed and dried on a sample placed approximately six inches from the nozzle. The water soluble polymer was poly(diallyldimethylammonium chloride) (Sigma-Aldrich, Mw ~1∙105 - 2∙105 g/mol).
Residue from the ultrapure process water was obtained by drying 2 μL droplets on a silicon wafer in a Class 100 clean room. The water was collected from a flowing stream with a polypropylene-tipped pipet. Prior to use, pipet tips were cleaned by an overnight soak in dilute hydrofluoric acid.
UV-O3 treatment was performed for 1 hour in a UV-O3 chamber (Jelight, Model 576). The O2 plasma treatment was performed for 1 hour in a plasma stripper (Branson/IPC PM-1813). Thermal treatments were performed for 2 hours in a benchtop furnace (Barnstead/Thermolyne Type 1400).
2.2 Laser damage tests
All laser damage tests were performed on the exit surface of the fused silica optics.
Small-area damage tests were performed with a 355 nm, ~80 μm diameter beam with a 3 ns Gaussian pulse. An imaging microscope was used to observe the sample during the damage test. The tests were performed using either the R/1 protocol (damage threshold reported by incrementally ramping the fluence) or the S/1 protocol (damage threshold reported as probability that a single exposure at a single fluence results in damage). The full system is described in detail elsewhere .
Large-area laser damage tests were acquired by illuminating the exit surface of a fused silica sample with a 355 nm, 1 cm diameter laser beam with a 5 ns flat-in-time pulse. The 5 ns flat-in-time pulse is roughly equivalent to the 3 ns Gaussian pulse used in the small-area damage tests from a damage initiation perspective . Mean exposure fluences were 30-40 J/cm2. Damage sites were imaged with an optical microscope and mapped to the measured spatial fluence distribution of the laser. The full system is described in detail elsewhere .
Optical microscopy of residue droplets was performed with a digital microscope (Keyence VHS2000 with Z100 lens or VIEW Benchmark 300). Detailed inspection was performed with a scanning electron microscope (SEM, FEI Nova600i Nanolab) operating at 5.0 kV accelerating voltage using a secondary electron detector. Qualitative elemental analysis was performed with energy-dispersive x-ray spectroscopy (EDX, EDAX Genius).
2.4 Water analysis
Impurities in the ultrapure water used in this study were analyzed by a combination of high-resolution inductively coupled plasma mass spectrometry (to measure inorganic species) and UV-persulfate oxidation (to measure total organic content). Analysis was performed by a commercial vendor (Balazs Nanoanalysis, Fremont CA, USA).
3. Results and discussion
3.1 Organic impurities in ultrapure water
The population of damage precursors is usually quantified by a large-area laser damage test, which consists of a laser exposure followed by quantitative measurement of the cumulative distribution of damage density as a function of fluence. These damage density curves have a characteristic shape with an asymptotic density at high fluences. This saturated damage density can be considered the total number density of damage precursors. In our previous report, we were able to reduce this saturated density by a factor of >100, but the shape of the damage density curve remained the same . Therefore, even at low damage densities (<103 cm−2), damage precursors are still likely to be discrete, uniformly distributed defects composed of foreign material from fabrication and processing. Any process step that introduces impurities is therefore a potential source of damage precursors. For example, immersion of a freshly etched and rinsed optic in an ultrapure water bath doped with <0.1% of untreated municipal water caused a measurable increase in damage density .
These observations suggest that trace impurities in the rinse water, even at very low levels, could be sufficient to create high-fluence damage precursors if they dried on the optic surface. Even ultrapure water that undergoes a thorough series of purifying steps (ion adsorption beds, UV-irradiation, particle filtration) contains trace impurities. Optics that dry in a vertical position (i.e., gravity assists in draining remaining rinse water from the optic face) may suffer from the deposition of these impurities as they concentrate at the receding drying front or may retain an ultrathin layer of water due to surface tension. In the latter case, even a nanoscopically-thin water layer may be problematic; a 5 nm thick water layer with 250 ng/L impurities can deposit nanoscale precursors (~150 nm in size) with an areal density of 103 cm−2.
While the quality of ultrapure water is measured by continuously monitoring its resistivity, this technique is limited to measuring charge-carrying impurities at levels greater than several ng/mL. It cannot detect ionic species below the ng/mL level nor can it detect contaminants that carry no charge, such as neutral organic species. Despite the use of UV illumination in the water purification system to sterilize and reduce the levels of organic impurities, their presence was observed by drying small aliquots of process water on a clean silicon surface. Figure 1 shows the residues that remained after the water evaporated. These residues have an amorphous, rather than crystalline, morphology and qualitative elemental analysis via EDX shows a substantial carbon signal. These observations suggest that the residue from the ultrapure process water is carbonaceous. The silicon signal from the analysis of the residue is due to the several-micron sampling depth of EDX. To rule out the possibility that the residue was partially siliceous, we performed a similar experiment on a germanium surface and found no silicon EDX signal in the residue (data not shown).
Analysis of the process water used in this work supports this conclusion. No ionic species were detected above their detection limits, which ranged from 0.5 to 50 pg/mL. The dominant impurities were organic species, with total organic carbon measured at a concentration of 350 ng/mL. Siliceous species were a minority impurity at 4.5 ng/mL. The actual identities of the organic impurities are unimportant both because a variety of organic compounds can form high-fluence precursors and because strategies to remove these precursors are insensitive to the bonding structure of any particular molecule, as will be subsequently discussed.
3.2 Damage behavior of organic impurities
Organic compounds such as phthalates (a common plasticizer) and sugars are known damage precursors when deposited on fused silica optics [11, 19, 20 ]. The early stages of the light absorption which leads to damage are not well-understood, but simple bulk absorption models do not predict the temperatures (>5000 K ) required to initiate a damage event . Moreover, a material need not be a bulk absorber to act as a damage precursor . Bien-Aimé and associates hypothesized a microlensing effect for spherical liquid droplets, but this mechanism could not account for the behavior of film-like organic liquids [19, 20 ]. Regardless of their absorption mechanism, carbonaceous contaminants on fused silica surfaces are problematic from laser damage perspective.
We simulated the drying of organic impurities on an optic surface by spray-deposition of a water-soluble polymer on an etched and rinsed fused silica substrate. The use of a water-soluble polymer simulated the formation of a discrete precursor from a dissolved carbonaceous impurity as might be found in ultrapure process water. The dried polymer formed discrete particles of <5 μm, many of which were sub-micron in size as seen in Fig. 2 .
Large-beam laser damage tests are best suited for measuring the density of precursors generated by a particular process. We wished to interrogate the damage behavior of a known material, rather than a fabrication process; therefore, we performed small-beam laser damage tests using the S/1 protocol. The small-beam damage test randomly sampled the surface rather than selectively testing visible particles. Optical microscopy, an example of which is shown Fig. 2, confirmed that laser damage sites were coincident with the site of one or more polymer particles prior to the damage test.
The small-beam damage tests found that a surface contaminated with these impurities had a markedly degraded damage threshold compared to an as-etched fused silica optic as shown in Fig. 3 . Precursors existing on the as-etched surface have a 50% damage probability of ~55 J/cm2. The precursors deposited by the polymer spray-deposition have a 50% damage probability of ~40 J/cm2.
These experiments show that a dissolved organic impurity can form discrete damage precursors when dried on fused silica. Trace organic impurities found in the ultrapure process water are therefore a potential source of laser damage precursors. Since any wet-processing will conclude in a rinse, the deposition of such impurities is practically inevitable even under stringent purity, cleanliness, and drying conditions.
3.3 Oxidative decomposition of organic damage precursors
Identifying organic impurities in process water as a source of damage precursors is important because mitigation strategies can be developed that are targeted specifically toward the removal of carbonaceous material. Ideally, the process should be solvent-free to prevent the redeposition of impurities carried by solvents, including water. Several simple dry strategies are based upon the oxidation of organic materials. Under completely oxidative conditions, carbon-based compounds decompose into volatile species dominated by H2O and CO2.
We examined three different oxidative environments in this work:
- UV-O3. The UV output of a mercury lamp breaks organic bonds and generates an atmosphere of highly oxidizing atomic oxygen.
- O2 plasma. Plasma fragmentation of O2 results in a variety of reactive oxygen species and the attendant vacuum ultraviolet light assists in breaking organic bonds.
- High temperature air. At high temperatures (>300 °C), thermal energy is sufficient to break organic bonds which then react with ambient O2. In this work, we restricted the thermal treatment temperatures to less than 90% of the glass transition temperature of fused silica in order to limit the introduction of mechanical stresses.
Rather than test these methods against model organic compounds, we instead used the organic impurities found in ultrapure water by drying small aliquots of process water on a clean silicon surface. Figure 4 shows optical micrographs of these residues before and after UV-O3 and O2 plasma treatments. Qualitatively, the O2 plasma treatment was more effective than the UV-O3 treatment. Thermal oxidation is also effective at removing residues. As shown in Fig. 5 , 500 °C for two hours in air reduced the amount of residue and no residue could be observed after an additional two hours at 990 °C.
We used each of these post-processes on freshly etched fused silica optics and characterized their laser damage behavior. Typically, damage density curves from high-fluence, large-area laser damage tests are used to measure the areal damage density. However, many of our samples had a small number of damage sites due to a combination of the low density of precursors and the limited beam size (~0.7 cm2). Poor counting statistics make it impossible to construct an accurate damage density curve. For such samples, we employed a different approach: because the density of damage sites saturates at high fluences, we instead counted the number of damage sites and assumed they occurred only over the area where the fluence was >35 J/cm2. This area was between 0.32 and 0.38 cm2 for all laser damage tests. This method provides an upper bound for the saturated damage density (i.e., the total number density of precursors) and facilitates comparison between various processing techniques.
The estimated saturated damage density from the large-area damage tests are shown in Fig. 6 . The as-etched fused silica results are the mean of eight samples from our previous report . While UV-O3 was only marginally effective, the other oxidative environments resulted in substantial reductions in the saturated damage density relative to the as-etched surface.
Thermal decomposition was the most effective post-processing technique, with estimated saturated damage densities of <50 cm−2 after both 500 °C and 990 °C treatments. Recall that these densities are upper-bound estimates due to the small number of observed damage sites (as low as 11 over the 0.7 cm2 beam area). The thermal decomposition results are consistent with laser heating experiments where a mid-IR laser is used to heat fused silica to high temperatures (~1700 °C) for several minutes in order to locally remove damage precursors. Unlike mid-IR laser heating, thermal decomposition in a furnace is a whole-optic rather than a local process. Additionally, mid-IR laser heating introduces local mechanical stress in the fused silica, whereas heating below the glass transition of fused silica (1200 °C) is not expected to introduce mechanical stress.
The effectiveness of the oxidative treatments tracks with their qualitative ranking from the residue experiments shown in Figs. 4 and 5 . Each of the treatments reduces the amount of residue by some degree, but UV-O3 is the least effective. This is consistent with the fact that this treatment reduces the saturated damage density somewhat but not substantially. On the other hand, the 990 °C thermal treatment left no observable residue and also had the lowest saturated damage density. These should be considered qualitative results useful for rank-ordering the various post-processing options. In practice, the reduction in the number of damage precursors will be a combination of the precursor size distribution and their decomposition rates under these conditions.
These techniques are ill-suited to mitigate inorganic contaminants because inorganic materials do not decompose to volatile species at these conditions. When attempted on a sample prepared prior to the implementation of protocols designed to reduce inorganic impurities, high-temperature treatments actually decreased the laser damage threshold of fused silica optics. Figure 7 shows the result of small-beam laser damage tests (S/1 protocol) on such an optic both before and after treatment at 1000 °C. Not only were damage precursors not removed, they became prone to damage at lower fluences.
Because the damage precursors on the as-etched sample used in this experiment were likely inorganic, they did not decompose under during the thermal treatment. Instead, the high temperatures likely increased the adhesion between the precursor and the silica surface.
To confirm this behavior, we placed grains of NaCl (~500 μm) on a silica surface and subjected them to conditions that increased their adhesion to the substrate: ambient humidity (40% RH), 55% RH, 90% RH, and 800 °C (the melting point of NaCl). As the quality of the adhesion increased, the NaCl grain needed to absorb more energy before it was be laser-cleaned from the surface as shown by the R/1 small-beam laser test in Fig. 8 . The damage threshold, based on S/1 small-beam laser tests, decreased with increasing adhesion.
Below the damage threshold, the grains absorbed enough energy to eject from the surface when the laser fluence was ramped as in the R/1 protocol, but did not remain on the surface long enough to initiate damage. As the adhesion increased, less energy was needed to initiate damage because the grain was less likely eject from the surface prior to the damage event, consistent with previous observations on metallic particles .
The cleaning and damage thresholds appear to converge with increasing adhesion, suggesting that a tightly adhered damage precursor will always initiate a damage event prior to being ejected from the surface. At very high temperatures even inorganic materials will decompose, as evidenced by the increase in damage threshold in areas heated to ~2000 K with a mid-IR laser [12, 13 ]; however, a whole-optic process above the glass transition temperature of fused silica is impractical.
The economical operation of high-fluence laser systems is a major driver for the continued improvement in the laser damage behavior of fused silica optics. Following the elimination of polishing residue, fractures, and inorganic contaminants, we have shown that a likely source for damage precursors is organic impurities in ultrapure process water. Dissolved organic compounds are the dominant impurity in the ultrapure water system used in this work. We have shown, using surrogate materials, that such materials can act as precursors when deposited on a fused silica surface.
As with inorganic materials, the details of the light absorption mechanisms which cause laser damage are not understood well for organic materials. Nevertheless, they represent an important class of precursors that can only be identified and mitigated when other sources of damage precursors have been suppressed to densities <103 cm−2. Dry methods to oxidatively decompose organic damage precursors include UV-O3, O2 plasma, and thermal treatments. Thermal treatment in air at 990 °C was the most effective technique observed in this work.
The combined suppression of inorganic impurities (demonstrated in an earlier report ) and decomposition of organic impurities (demonstrated here) has reduced the total damage density on fused silica by a factor of 2000 . In addition to reducing the saturated damage density, such advances to processing will also enable UV lasers to operate at higher fluences without damaging fused silica optics.
The ability of these treatments to reduce the saturated damage density demonstrates that organic damage precursors are a limiting factor for high-quality fused silica optics. In the presence of inorganic damage precursors which do not decompose oxidatively, high temperature treatments serve to increase their adhesion to the substrate and make them more damage-prone. These methods treat the whole optic without introducing additional mechanical stresses and may be scaled for large-aperture optics. Understanding the precursor size distribution and their etch rates under various oxidation conditions will be an important aspect of future work. Such measurements would allow for predictive models when engineering oxidative decomposition processes for organic damage precursors.
Authors acknowledge the assistance of N. Teslich (SEM and EDX analyses); D. VanBlarcom (UV-O3 treatment); J. Hayes and C. Alford (O2 plasma treatment); and W. Carr, M. Norton, and D. Cross for large-area damage testing. This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344 within the LDRD program.
References and links
1. L. B. Glebov, “Intrinsic laser-induced breakdown of silicate glasses,” Proc. SPIE 4679, 321–331 (2002). [CrossRef]
2. S. C. Jones, P. Braunlich, R. T. Casper, X.-A. Shen, and P. Kelly, “Recent progress on laser-induced modifications and intrinsic bulk damage of wide-gap optical materials,” Opt. Eng. 28(10), 1039–1068 (1989). [CrossRef]
4. C. W. Carr, J. D. Bude, and P. DeMange, “Laser-supported solid-state absorption fronts in silica,” Phys. Rev. B 82(18), 184304 (2010). [CrossRef]
6. C. A. Haynam, P. J. Wegner, J. M. Auerbach, M. W. Bowers, S. N. Dixit, G. V. Erbert, G. M. Heestand, M. A. Henesian, M. R. Hermann, K. S. Jancaitis, K. R. Manes, C. D. Marshall, N. C. Mehta, J. Menapace, E. Moses, J. R. Murray, M. C. Nostrand, C. D. Orth, R. Patterson, R. A. Sacks, M. J. Shaw, M. Spaeth, S. B. Sutton, W. H. Williams, C. C. Widmayer, R. K. White, S. T. Yang, and B. M. Van Wonterghem, “National Ignition Facility laser performance status,” Appl. Opt. 46(16), 3276–3303 (2007). [CrossRef] [PubMed]
7. P. E. Miller, J. D. Bude, T. I. Suratwala, N. Shen, T. A. Laurence, W. A. Steele, J. Menapace, M. D. Feit, and L. L. Wong, “Fracture-induced subbandgap absorption as a precursor to optical damage on fused silica surfaces,” Opt. Lett. 35(16), 2702–2704 (2010). [CrossRef] [PubMed]
8. P. E. Miller, T. I. Suratwala, J. D. Bude, T. A. Laurence, N. Shen, W. A. Steele, M. D. Feit, J. A. Menapace, and L. L. Wong, “Laser damage precursors in fused silica,” Proc. SPIE 7504, 75040X (2009). [CrossRef]
9. 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]
10. T. I. Suratwala, P. E. Miller, J. D. Bude, W. A. Steele, N. Shen, M. V. Monticelli, M. D. Feit, T. A. Laurence, M. A. Norton, C. W. Carr, and L. L. Wong, “HF-Based Etching Processes for Improving Laser Damage Resistance of Fused Silica Optical Surfaces,” J. Am. Ceram. Soc. 94(2), 416–428 (2011). [CrossRef]
11. J. Bude, P. Miller, S. Baxamusa, N. Shen, T. Laurence, W. Steele, T. Suratwala, L. Wong, W. Carr, D. Cross, and M. Monticelli, “High fluence laser damage precursors and their mitigation in fused silica,” Opt. Express 22(5), 5839–5851 (2014). [CrossRef] [PubMed]
12. R. M. Brusasco, B. M. Penetrante, J. A. Butler, S. M. Maricle, and J. E. Peterson, “CO2-laser polishing for reduction of 351-nm surface damage initiation in fused silica,” in Laser-Induced Damage in Optical Materials: 2001 Proceedings, G. Exarhos, A. H. Guenther, K. L. Lewis, M. J. Soileau, and C. J. Stolz, eds. (2002), pp. 34–39.
13. R. N. Raman, M. J. Matthews, J. J. Adams, and S. G. Demos, “Monitoring annealing via CO2 laser heating of defect populations on fused silica surfaces using photoluminescence microscopy,” Opt. Express 18(14), 15207–15215 (2010). [CrossRef] [PubMed]
14. R. M. Vignes, T. F. Soules, J. S. Stolken, R. R. Settgast, S. Elhadj, and M. J. Matthews, “Thermomechanical Modeling of Laser-Induced Structural Relaxation and Deformation of Glass: Volume Changes in Fused Silica at High Temperatures,” J. Am. Ceram. Soc. 96(1), 137–145 (2013). [CrossRef]
15. S. T. Yang, M. J. Matthews, S. Elhadj, V. G. Draggoo, and S. E. Bisson, “Thermal transport in CO2 laser irradiated fused silica: In situ measurements and analysis,” J. Appl. Phys. 106(10), 103106 (2009). [CrossRef]
16. N. Shen, P. E. Miller, J. D. Bude, T. A. Laurence, T. I. Suratwala, W. A. Steele, M. D. Feit, and L. L. Wong, “Thermal annealing of laser damage precursors on fused silica surfaces,” Opt. Eng. 51(12), 121817 (2012). [CrossRef]
17. C. W. Carr, D. A. Cross, M. A. Norton, and R. A. Negres, “The effect of laser pulse shape and duration on the size at which damage sites initiate and the implications to subsequent repair,” Opt. Express 19(S4Suppl 4), A859–A864 (2011). [CrossRef] [PubMed]
18. C. W. Carr, M. D. Feit, M. C. Nostrand, and J. J. Adams, “Techniques for qualitative and quantitative measurement of aspects of laser-induced damage important for laser beam propagation,” Meas. Sci. Technol. 17(7), 1958–1962 (2006). [CrossRef]
19. K. Bien-Aimé, C. Belin, L. Gallais, P. Grua, E. Fargin, J. Néauport, and I. Tovena-Pecault, “Impact of storage induced outgassing organic contamination on laser induced damage of silica optics at 351 nm,” Opt. Express 17(21), 18703–18713 (2009). [CrossRef] [PubMed]
20. K. Bien-Aimé, J. Néauport, I. Tovena-Pecault, E. Fargin, C. Labrugère, C. Belin, and M. Couzi, “Laser induced damage of fused silica polished optics due to a droplet forming organic contaminant,” Appl. Opt. 48(12), 2228–2235 (2009). [CrossRef] [PubMed]
21. T. A. Laurence, J. D. Bude, S. Ly, N. Shen, and M. D. Feit, “Extracting the distribution of laser damage precursors on fused silica surfaces for 351 nm, 3 ns laser pulses at high fluences (20-150 J/cm2),” Opt. Express 20(10), 11561–11573 (2012). [CrossRef] [PubMed]