1. Introduction

Advances in bulk purification and surface finishing methods for optical materials [1,2], have significantly enhanced the damage resistance of optical components used in high power laser systems. However, laser-induced damage continues to limit operational lifetime of most components via obscuration of the clear aperture. The initial size of such damage sites is highly dependent on the laser parameters [3,4] but ranges from a few microns to a few tens of microns for nanosecond pulses. As initiated, these sites are sufficiently sparse that they would pose no significant threat to beam throughput were it not for their propensity to grow exponentially in diameter upon exposure to subsequent pulses (damage growth) [5–7]. Thus, the study of the nature of damage growth is of great interest in terms of 1) understanding the fundamental response of wide-bandgap dielectrics in general – and optical materials in particular – to laser energy deposition and 2) impeding and possibly arresting such growth behavior. In working towards these goals, it has been observed that a number of parameters affect the rate at which sites grow – laser fluence, pulse duration, and wavelength, as well as site morphology. Moreover, the rate at which sites with nominally identical diameter exposed to nearly identical laser pulses grow is not deterministic. The site diameter and laser parameters do govern the distribution of growth rates in a repeatable and predictable manner [7–9]. This observation requires that additional innate characteristics of the site, apart from surface diameter, determine where an individual site will reside in a distribution of growth rates. To date it is unknown which damage site attributes lead to more aggressive growth; however, it has been observed that individual damage sites with identical diameters can substantially differ morphologically [7,10,11]. Furthermore, previous work has found an association between structural properties and propagation of flaws in glass [12–14], suggesting such structural differences may also relate to how fast a site will grow upon subsequent laser exposure.

Structural properties of ceramic materials (e.g., ZnO, Al₂O₃, and MgO) have been traditionally probed by thermal anneal [15]. More recently, thermal anneal has been shown to improve optical transmission and laser-induced damage threshold of materials as diverse as ZnSe crystals [16], potassium dihydrogen phosphate [17], oxide thin films [18], and fused silica [19,20]. A study on the isothermal anneal of cracks created by mechanical indentation in silica showed partial crack healing at temperatures as low as 525°C in the form of crack pinching and crack tip blunting driven by capillary flow [12]; increasing the anneal temperature and time revealed a treatment regime of increasing mechanical strength until a maximum was reached, followed by a decline.

Thermal anneal of laser-induced damage on SiO₂ surface has also been achieved by exposure to 10.6 μm CO₂ laser irradiation (SiO₂ absorption coefficient ~10³ cm⁻¹), leading to significant surface smoothing and an arrest of growth [21,22]. CO₂ laser heating has the advantage of permitting the observation of local, in situ healing of damage sites [23] as a function of various treatment parameters; however, identical and simultaneous treatment of a large number of sites is not feasible.

In this work, we investigate the effect of an oven anneal on growth rate (defined here as the rate of growth laterally) of both small (as-initiated, <50 μm diameter) and large (grown, >100 μm diameter) damage sites in fused silica. Furthermore, we explore the relationship between growth rate and measurable anneal-induced changes in surface and sub-surface damage structure, relative concentration of ultraviolet (UV) light absorbing species, and stress-induced birefringence.

2. Experimental procedure

2.1 Sample preparation

Two UV-grade fused silica flat rounds (Corning 7980 Type III glass, ~1000 ppm OH content by wt. % [24]. were used in this study. Sample A, intended for studying damage growth using a large aperture beam, was 1 cm in thickness and 2 inches in diameter. Sample B, intended for side-viewing of damage site features extending from the surface into the bulk, began as an identical cylinder and was then cut to produce a 1-cm thick square with rounded corners. All four edges were polished to allow imaging through the sides of the sample. Both samples were then exposed to a chemical etchant to improve surface resistance to laser damage [2].

Damage sites were initiated on the bare surface using a single laser pulse from a 355 nm, 3.5-ns (measured as Full Width at Half Maximum of intensity, FWHM) Nd:YAG Q-switched laser (EKSPLA). The laser beam with energy of ~1 mJ was focused down to a ~50 μm diameter spot at the exit surface of the samples using a 300 mm focal lens. The laser fluence at the sample’s surface exceeded the surface damage threshold (~60 J/cm²) and resulted in damage with ~100% probability. The damage manifested itself mostly as single pits with a diameter of 40 ± 10 μm (mean ± standard deviation). Each sample was translated to expose a pristine location to each subsequent pulse. For Sample A, a grid of damage sites was generated across the surface within a 20 mm x 20 mm region of interest centered on the round [25]. To establish an internal control, initially only 5 rows of damage sites (55 sites with 2 mm nearest neighbor spacing) were created on the lower half of this region of interest (designated as Set 1 or control). In this manner, site growth characteristics before the anneal treatment could be determined. For Sample B, one row of damage sites was initiated 5 mm from each edge, yielding a total of 40 sites with 2 mm nearest neighbor spacing.

Following initiation using the small laser beam, the damage sites on Sample A were simultaneously grown using a large aperture (~30 mm) laser beam at the Optical Sciences Laboratory at LLNL. This laser, described in detail elsewhere [26] is a Nd:glass amplifier laser system outputting a 3rd harmonic, 100 J pulse with tunable temporal width and shape. Damage growth proceeded in near-vacuum (<10⁻⁶ Torr) and at room temperature by exposing the sample to a series of five, nearly identical laser pulses at 351-nm, 10.0 ± 2.0 J/cm² (mean ± standard deviation), 5-ns flat-in-time (FIT). For each growth shot, the spatial distribution of the fluence within the large aperture beam was recorded and subsequently registered to the sample, providing the local fluence at each site (standard error ~1% of the mean value obtained by averaging over a 500 μm x 500 μm square patch). Between laser exposures, the sites were characterized with a number of methods, as described in Section 2.2. Upon completion of the damage growth sequence and site characterization of Set 1, additional 55 sites were initiated in the remaining half of the region of interest (designated as Set 2). Following thermal anneal of Sample A (procedure described below), sites from both Set 1 and Set 2 were then simultaneously exposed to five additional laser shots at the same fluence (10.0 ± 2.0 J/cm²) to determine their growth rate.

The thermal anneal procedure consisted of heating the samples in an oven at a rate of 10°C/minute from room temperature to 1100°C, then an isothermal hold for 12 hours with steady-state fluctuation of ± 1°C, followed by a 0.5°C/minute cool down back to room temperature. This peak temperature was chosen to slightly exceed the annealing point of type III silica at 1042°C [24], the temperature at which the shear relaxation time $\tau =\eta (T)/G$ = ~30 sec for viscosity η = 10¹² Pa∙s and the shear modulus G ~30 GPa [27,28].

2.2 Diagnostic imaging

After each laser exposure of the damage growth protocol, top-view optical micrographs of each site were acquired using a robotic microscope under back-illumination (LED emission centered at 532 nm) and oblique-illumination (white light). A field of view of 680 x 510 μm and ~1 μm resolution (Benchmark 200, View Microscopy and Metrology) is achieved using a 20X objective with 0.42 NA, 20 mm working distance, and 1.6 μm depth of field (ELWD 20X, Motic). From these images, the effective circular diameter (ECD) and other morphological features of the sites were measured. The ECD is defined as the diameter of a circle with area equal to that of the damage site and is inferred by thresholding the image beyond the background intensity level.

Each damage site on Sample B (the side cut-and-polished sample) was imaged through the nearest polished side. Similar light settings as those for top-view measurements were employed for side-view imaging. Images were captured by the robotic microscope both before and after anneal in order to determine the sub-surface morphological response of damage sites.

Scanning electron microscopy (SEM) images of sites were also captured before and after the thermal anneal in order to record sub-micron changes on the damage site surface. The SEM instrument (Hitachi S3400-N) was set to use 2.5kV electrons and an emission current of 70 μA to image sites in vacuum (<10⁻⁵ Torr), with spatial resolution on the order of 0.1μm. To reduce charging by the electron beam, images were acquired by averaging 32 short exposures for a total exposure time of ~1 min per site. A separate control experiment found no measurable effect of electron beam exposure at this (or any) level on the growth rate.

The robotic microscope used a polarized ~530 ± 35 nm light source. The light transmitted through the sample was filtered with a linear polarizer oriented perpendicular to the input polarization state. This geometry allows imaging of stress-induced birefringence in a dark-field configuration for increased sensitivity, i.e., no signal when the surface and bulk of a sample are pristine. Depolarized light reaches the detector by either 1) scattering off of multiple surfaces where the ensemble average polarization state is randomized (as can occur when light is scattered by the complex morphology of the damage site) or 2) stress-induced birefringence (as can occur in the volume beneath the core or surrounding the damage site). We will primarily focus our discussion on the latter source of depolarization as observed at the damage site periphery where the scattering contribution is negligible. The difference in the principal stress components,$\Delta \sigma$, normal to the direction of transmitted light propagation can be estimated by

Photoluminescence (PL) measurements were performed using a PL microscope apparatus previously described [29]. In brief, sites were exposed to 355 nm laser illumination (1 mW) off-axis for 30 s while the 420 nm long-pass filtered emission was imaged with resolution on the order of 2 µm onto a liquid nitrogen-cooled CCD camera. This emission is known to originate primarily from the following electronic defects in fused silica: non-bridging oxygen hole centers (NBOHC, peak emission at 650 nm) and laser-induced defects (LID, peak emission at 560 nm) [30,31]. The mean PL intensity count rate (ratio of the mean PL counts to UV laser exposure time) from each image was calculated over a circular region of interest with diameter equal to the damage site ECD.

2.3 Determination of growth rate and growth probability

Growth was defined as an event when the site’s effective circular diameter ECD_n after shot n exceeded its pre-shot diameter ECD_n-1 by at least 2 µm. This criterion is derived from the resolution of the measurement of site diameter using our imaging system [7] and is valid for sites with ECD >5 μm. In this manner, multiple shots on the same site were treated as independent events in order to generate a population size adequate for reliable statistical analysis of growth rate (this treatment is supported in [7]). For example 55 sites were grown with 5 large aperture shots, resulting in 55 x 5 = 275 growth events. Since in [7] it was found that growth rate is influenced by ECD and fluence, growth events were binned in these categories: 30-50, 50-70, 70-100, and 100-200 µm size bins for ECD and 10 ± 1 J/cm² for local fluence. Exposures with fluence outside of the 9-11 J/cm² bin were discarded from analysis since sample sizes were insufficient for reliable statistical analysis. Single-shot growth rate (α) was calculated based on the expectation of exponential increase in ECD upon exposure to multiple, 5-ns duration laser pulses [7]:

A reliable estimation of the likelihood that a population of sites grows (probability of growth) requires analysis of a larger sample size and, for a general description, sampling over a wider fluence range [9]. Nonetheless, probability of growth was estimated in this study as the fraction of growing sites of comparable diameter and laser exposure fluence binned as described above). Here, due to the coarser binning by size, we effectively reduce the influence of site diameter on the probability of growth as quantified in [9], where it was found that smaller sites have distinctly lower probability of growth compared to larger sites for a given 351 nm laser fluence.

3. Results

All sites in the unannealed population grew on the first shot, while 2 out of 43 annealed sites (~5%) did not grow. These non-growing sites (α = 0) were excluded from analysis. The thermal anneal process alone did not significantly change the ECD of either the small (as- initiated) or large (grown) sites; when the ECD distribution was compared before and after anneal, the p-value (probability of achieving the observed difference in ECD from identical populations by random sampling) was >>1% (0.70 and 0.15, respectively, using unpaired t-test). Thus any changes in ECD are attributed to damage growth.

Figure 1(a) shows the progression of the growth rate of the control (unannealed) site population (Sample A, Set 1) as well as that of the annealed site population (Sample A, Set 2) vs. shot number. Since site size varies from shot to shot, the ECD bin is noted for each shot in Figs. 1(a)-1(b). Before comparing the growth behavior of annealed vs. control sites, a comment on size bin is warranted. From shot 2 onwards for both populations, a progressive decrease in growth rate (independent of anneal) is observed; previous work has identified that as the mean ECD gets progressively larger with each laser shot, growth rate is reduced [8].

 

Fig. 1 (a) Shot-by-shot progression of growth rate (mean ± standard error) of damage sites under 10 ± 1 J/cm², 351 nm laser fluence annealed at 1100°C (circles) compared to that of unannealed sites (squares). (b) Progression of growth rate of large sites (100-200 μm ECD) before and after anneal. Site diameter range and p-values are indicated.

Download Full Size | PDF

For the first shot, small annealed sites exhibited a statistically significant reduction in growth rate (p<0.0001, unpaired t-test), from 0.40 ± 0.11 compared to that of unannealed sites 0.23 ± 0.09 (mean ± standard deviation); this difference in growth rate remained significant but was reduced on the second shot. By the third shot, there was no statistically significant difference between mean growth rates of annealed and control sites (p>0.01).

Control sites with ECD 100-200 μm were growing with α = 0.23 ± 0.07 compared to α = 0.18 ± 0.03 immediately after anneal (demarcated by the red line, Fig. 1(b)); while there is a reduction in the mean growth rate of sites before vs. after anneal, this difference is much less than that for small sites. Part of this reduction in Fig. 1(b) is anneal-independent and is due to the departure from the bin of the fastest growing sites, as observed in the downward trend of shots 3-5. However, this observation represents the maximum possible contribution due to anneal, which is still less than that for small sites. Another result is that the standard deviation of the α distribution decreased upon anneal (from 0.11 to 0.09 for small sites and from 0.07 to 0.03 for large sites).

For the remainder of this section, we examine several site attributes inferred from diagnostic imaging (Section 2.2) that may explain the difference in growth rate response to thermal anneal observed in small vs. large sites. Figure 2 shows typical polarimetry images captured before and after anneal for two damage sites with diameters of 175 μm (a-b) and 40 μm (c-d). These images (linear intensity scale) were contrast-enhanced to highlight the stressed peripheral regions. The larger damage site was part of Set 1, grown with 5 laser shots at ~10 J/cm². Figures 2(a)-2(b) indicate that the signal coming from the damage site core issomewhat reduced, while the signal at the site periphery is almost extinct after anneal. We estimate the maximum normalized stress difference at the location indicated by arrow in Fig. 2(a) to be ~0.45 ± 0.10 MPa/cm⁻¹. The uniform, near extinct signal at the periphery upon anneal suggests almost complete stress relaxation in the bulk material surrounding the large damage site.

 

Fig. 2 Response of stress-induced birefringence to anneal: (a-b) are images of a large damage site, while (c-d) are images of a small damage site. Arrows point to regions of maximum birefringence outside of the fractured region.

Download Full Size | PDF

The small site in Figs. 2(c)-2(d) is representative of as-initiated damage sites in Set 2. While little stress was detected with this microscope in such cases, a reduction in signal from beneath the damage site was still resolvable after anneal (Figs. 2(c)-2(d)). Thus, both large and small sites alike did not exhibit any significant amount of stress-induced birefringence following anneal. This result is by design, as the oven temperature was chosen to exceed the glass annealing temperature and permit local stress relaxation, while cooling down slowly enough to avoid quenching in a higher fictive temperature and introducing additional stress.

Figure 3 displays typical photoluminescence images of (a-b) a large (ECD ~290 µm) and (c-d) a small site (ECD ~40 µm) on a pseudo-color intensity scale before and after anneal. Larger sites exhibited a greater mean PL intensity, likely due to either a greater density of electronic defects or greater volume of material containing these defects. For both large and small sites, the PL signal from electronic defects (absorbing 355 nm radiation) was significantly reduced (80% reduction for large sites, 70% reduction for small sites) following anneal. The reduction in the PL signal is likely indicative of reduction in the concentration of electronic defects, expected at anneal temperatures as low as a few hundred degrees Celsius and found in the damage site core as well as at cracked surfaces [19,32–34].

 

Fig. 3 Response of photoluminescence to anneal (pseudo-color) under 355 nm excitation. (a-b) are images of a grown damage site, (c-d) are images of a small damage site.

Download Full Size | PDF

The SEM images of Fig. 4 show the response of both a large and a small damage site to anneal. Fibers believed to have formed from material liquefied by the high-temperatures [35] during laser-induced-damage are observed to smooth upon anneal (an example is indicated by arrow in Fig. 4(a)). In addition, 100 nm-wide striations known as hackle can be observed to smooth upon anneal (see arrow, Fig. 4(c)). These SEM images acquired with 0.1 μm resolution suggest that features on the crater surface larger than ~150 nm remain unsmoothed following the anneal.

 

Fig. 4 Comparison of SEM images before and after anneal for a large (a-b) and a small (c-d) site. Top arrow (a) indicates a nanofiber, while bottom arrow (c) indicates hackle.

Download Full Size | PDF

Figures 5(a)-5(f) show the response of light scattering from small damage sites to anneal under back- and oblique-light scattering configurations. Top-view backlit (Figs. 5(a)-5(b)) reveal the disappearance of surface radial cracks up to 3 µm thick and 15 µm long, thinning of wider cracks, and a recession in the radius of subsurface fracture. In addition, the annealprocess induced a ~14% decrease in oblique light scattering (Figs. 5(c)-5(d)). This reduction could be due to the smoothing of the crater surface rubble as well as to the scattering from subsurface voids and cracks. However, images from SEM have confirmed that no significant smoothing of the crater occurs, and the size of features that were smoothed (<150 nm) are too small to efficiently scatter visible light. Thus it is more probable that a reduced amount of sub-surface fracture is responsible for the reduced light scattering. Figures 5(e)-5(h) (backlit side-view images) reveal the response of the subsurface crack network to anneal. One can identify nearly 100% closure of subsurface lateral cracks (as measured by the percent decrease of dark pixels under the crater) in the bulk up to 9 µm thick and 45 µm long. However, these images also reveal incomplete closure (45% reduction) of subsurface fracture as seen in the large sites (Figs. 5(g)-5(h)).

 

Fig. 5 Top-view (a-d) and side-view (e-h) images showing light scattering response to anneal for sites with different ECD. All images are backlit with the exception of (c-d), which are lit off-axis.

Download Full Size | PDF

4. Discussion

In this study we generated two populations of damage sites which responded differently in their growth behavior to thermal anneal at 1100°C for 12 hours; freshly initiated (small) sites exhibited a noticeable reduction in growth rate, whereas large sites grown with several laser exposures exhibited a much smaller reduction in growth rate (see Fig. 1). Moreover, anneal narrowed the α distributions for both populations meaning they grew more predictably. A number of mechanisms have been proposed for damage site growth: namely residual stress which drives crack propagation [13], the sub-bandgap absorption of the incident pulse energy by electronic defects [30], field intensification brought about by surface roughness [36], and reduced mechanical strength due to the presence of cracks [13]. Next, we examine the contributions of these various phenomena in small (ECD~40 μm) versus large (ECD~100-200 μm) sites in order to elucidate the difference in growth rate reduction following anneal.

4.1 Residual stress

The choice of 1100°C for the hold temperature of the anneal protocol allowed selective interrogation of these features believed to be responsible for laser-induced growth. As this temperature resides above the glass transition temperature for type III silica glass, any residual stress in the damage sites was allowed to locally relax. For example, the expected reduction in stress $\sigma /{\sigma }_0$due to shear relaxation for a 12-hour, 1100°C anneal is _{$Exp(t/\tau )$} ~0 for t = 12 hrs and $\tau$ = 0.005 hrs. The substantial reduction in stress-induced birefringence in both small and large sites (Fig. 2) confirms this response, and therefore does not explain the difference in growth reduction after thermal anneal. Additionally, we consider that the anneal probably depleted some of the OH content at the surface, rendering the volume surrounding the surface-bound region of the damage site stiffer than the volume located further into the bulk. This difference in viscosity could also provide a difference in the amount of stress relaxation around the cracks of small vs. large, deeper sites. Assuming a substantial reduction in [OH] down to ~3 ppm wt. % and estimating the corresponding viscosity (3.9 x 10¹³ Pa∙s) [27], the stress would still substantially relax to $\sigma /{\sigma }_0$~0 for t = 12 hrs and _$\tau$ = 0.42 hrs.; due to the long anneal hold time, the effect of OH depletion is negligible.

4.2 Photoluminescence

Point defects allow for a possible mechanism of energy transfer from the sub-bandgap laser pulse to the silica. Annealing of these defects has been well documented to be driven primarily by diffusion and has been observed to occur efficiently at temperatures well below that employed in this study (~500°C or less) [19,32,33]. In the present work, a 12 hour anneal at 1100°C dramatically reduced the photoluminescence signal, which indicates that the concentration of defects capable of absorbing UV light is reduced (though not eliminated) in both small and large sites. This common response in PL emission intensity to the anneal suggests that the presence of absorbing defects is not responsible for the observed dissimilarity in growth rate for these 2 populations following anneal.

4.3 Surface smoothing

We turn now to analysis of the morphological changes of damage sites observed in this study. An estimate of the size of features expected to be smoothed due to capillary-driven relaxation can be made (neglecting diffusion and evaporation) using the relation$L~\frac{\gamma \pi \tau }{\eta }$ [37], where L is the feature length, γ the surface tension, η the viscosity (a function of temperature and time), and τ the relaxation time (taken here as the anneal hold time). Using published values for γ (0.3 N/m) [38] and η (5.5 x 10¹¹ Pa∙s) at 1100°C [27] for type III fused silica, L is estimated to be of the order of 100 nm. This value is close to the scale of smoothing observed in both small and large sites (<150 nm, Fig. 4). Field intensification is unlikely brought about by features smaller than the laser wavelength (355 nm) where Rayleigh scattering dominates and does not explain the ECD-dependent reduction in growth rate following anneal.

4.4 Crack closure

The scale of features smoothed in cracks (tens of microns) is larger than the scale of smoothing observed in the surface morphology (<100 nm). The reason for this difference in response is that surface smoothing in glass is driven purely by capillarity-driven viscous flow, whereas crack closure is driven by residual strain in addition to viscous flow. When a crack is formed, the surrounding material is rendered in a highly strained and densified state. We hypothesize that when the glass is heated, this strain begins to relax. This additional driving force, along with three-dimensional thermal expansion which itself closes the physical gap of a crack, is what is responsible for the increased scale of smoothing observed in crack closure.

Cracks are associated with both mechanical strength and presence of electronic defects [34]. Since the PL measurement largely discounted absorbers as the main cause of the anneal-induced response to growth rate, we focus on the influence of the crack network on a site’s mechanical strength. Crack healing in sapphire crystals has been proposed to occur by blunting of crack tips forming cylindrical voids, then subsequently into spherical pores [39]. In that study, it was shown that crack healing was largely driven by surface and volume diffusion, and in related studies it has been shown that thermal healing of these cracks returned the fracture strength of the ceramic to its state prior to thermal shock [15]. Since the cracks in this study were not observed in real-time during anneal, the exact mechanism by which they thermally heal in fused silica was not determined. However, we note that the previously observed transition between the strengthening and weakening regime for indented glass corresponded to a value of t/τ ~10⁵ [12]; in the present study our anneal protocol yields t/τ ~10³, well within the strengthening regime. Thus, the significant decrease in the growth rate following anneal is associated with an increase in fracture toughness of the small sites. We hypothesize that the thermally-driven closure of cracks leaves the site in a more plastically deformed (less brittle) state, permitting the bulk to absorb more of the energy transferred from the laser pulse. This in turn reduces the amount of subsequent fracture (reduced overall site size after a laser exposure, hence lower growth rate) than would arise had these cracks not been healed. Figure 5 illustrates how this effect is more pronounced in the smaller sites by the healing of thinner, narrower cracks readily achievable by capillary flow at this anneal temperature; the larger sites instead have a much more complex fracture structure which frustrates relaxation. This effect is short-lived, as two more growth shots render the annealed site with a growth rate as if there were no anneal; it is possible that at this stage, the damaged volume has been ejected and fresh cracks are introduced as the site grows. Further studies tracking the evolution of the crack network within the bulk is necessary to validate this hypothesis.

5. Summary

In conclusion, this study demonstrated a reduction in the growth rate of annealed damage sites, more substantial for small sites than for large sites. Large sites did not respond any differently than small sites with respect to relaxation of stress, electronic defects, and susceptibility to downstream intensification (Table 1). Rather, large sites contained cracks too wide to heal under the present protocol. We conclude that residual fractures render these sites structurally weaker than small sites and thus prone to larger growth events. At even higher temperatures it is expected that the material will relax under capillary flow, fusing these wider cracks.

Table 1. Summary of Findings

View Table