Evolution of the point defects involved under the action of mechanical forces on mechanically machined fused silica surfaces

Dinghuai Yang; Jian Cheng; Linjie Zhao; Mingjun Chen; Henan Liu; Jinghe Wang; Chengshun Han; Zhichao Liu; Shengfei Wang; Feng Geng; Yazhou Sun; Qiao Xu

doi:10.1364/OE.483756

1. Introduction

Due to its excellent optical properties, fused silica has been widely used in many occasions including lenses [1] and shields [1] in high-power laser systems. Recently, to meet the ever-growing demands of the fields of high-power-laser technology for higher output power, such as ultrashort pulse laser processing, laser weapons, extreme ultraviolet lithography and laser-driven inertial confinement fusion (ICF) facilities, the laser output powers have increased sharply, putting forward higher requirements for the fused silica optical components with high-quality surfaces [2–4]. However, due to the fragile properties of fused silica [5], plenty of surface defects, like pits, cracks and scratches are inevitably induced on mechanically machined optical surfaces [6,7]. Surface defects would greatly increase the laser absorption of optical components and so sharply reduce the laser damage resistance under intense laser irradiation [8].

It has been widely accepted that point defects would be introduced in surface-defect zones under the actions of mechanical forces [9]. Previous work indicates that point defects could play as laser-damage precursors [10]. Moreover, they have been regarded as the main factor of increasing the laser absorption and reducing the laser damage resistance of the mechanically machined optical surfaces [10]. It has also been widely accepted that the point-defect densities in surface-defect zones are closely related to the laser-induced damage thresholds (LIDTs) of the mechanically machined optical surfaces [11,12]. Chen et al. found that there is a inverse proportional correlation between the PL induced by the point defects gathered in the surface defects and the LIDTs of the surface defects [13]. Recently, alleviating the point defects gathered in the surface defects to improve the laser damage resistance of the mechanically machined fused silica optical surfaces has been a research hotspot. Li et al. used the inert ion beams with large incident angle for layer by laser etching of the mechanically machined fused silica surfaces [14]. They found that the decrease of the structural defects (point defects) could significantly improve the LIDT of the mechanically machined fused silica optical surfaces. Shen et al. found that the densities of the point defects located on the SiO₂ films would decrease in the annealing process of the SiO₂ films, which distinctly improves the LIDTs of the SiO₂ films [15]. Besides, some work used the UV laser conditioning method to alleviate the point defects gathered in the surface defects to improve the laser damage resistance of the mechanically machined fused silica optical surfaces [16]. However, the systematical and quantitative study on the point defects located on the mechanically machined fused silica surfaces was still rarely reported, which restricts the further development of these methods to improve the laser damage resistance of the fused silica optical surfaces. For example, if the point defects with the largest proportion or the greatest impact on the laser damage resistance can be obtained, researchers can use these methods to target the removal of these point defects to maximize the increase of the LIDTs of the mechanically machined fused silica surfaces. Moreover, previous work [17,18] indicates that the point defects located on the mechanically machined fused silica surfaces could induce sub-bandgap defect energy levels, which could significantly increase the laser absorption of the material and resultantly sharply decrease the LIDTs of the mechanically machined surfaces. It is also considered to be the reason why the LIDT of the mechanically machined fused silica surface is far lower than its theoretical value (e.g., ∼100J/cm² @3 ns, 355 nm). However, the comprehensive and systematic study on the point defects was still rarely reported. Therefore, exploring the point defects located on the mechanically machined fused silica optical surfaces is of profound significance. Especially, various point defects with different electronic structures and optical properties are located in surface-defect zones [19]. While the effects of various point defects on the laser damage resistance of the mechanically machined optical surfaces are distinctly different [20]. Previous work has put efforts in exploring the point defects with various species on fused silica surfaces. Skuja et al. detected a direct singlet-to-triplet optical absorption transition (S₀ to T₁) in the twofold-coordinated silicon center (‘B₂(Si) center’) by time-resolved photoluminescence (PL) techniques [21]. Skuja et al. discussed the possible interconversion mechanisms between 2-fold-coordinated Si, neutral oxygen vacancies and E’-Center [22]. However, the study on the point defects induced by the mechanical forces was still reported less. Thus, it is significant to systematically explore the origins, the evolution laws of the point defects on the mechanically machined optical surfaces and especially the quantitative relationship among various point defects to quantitatively explore the comprehensive effect of various point defects on the laser damage resistance of the mechanically machined optical surfaces, which has been rarely reported in previous work.

Recently, photoluminescence (PL)-detection method, electron paramagnetic resonance (EPR) method, Fourier-transform infrared spectrometry (FTIR) method and Spectrophotometry method have been considered as the mainstream detection methods to characterize the point defects located on optical surfaces [19,23,24,25]. These detection methods characterize point defects by analyzing the characteristic signals closely related to the point defects [19,20,24]. For PL-detection, FTIR and Spectrophotometry methods, although these methods could effectively determine the point-defect types on mechanically machined optical surfaces through the characteristic signals of the point defects, they cannot obtain the proportions and the densities of the point defects due to the lack of knowledge on the relationship between the characteristic-signal strength and the point-defect density [19,20,24,26]. Due to the theoretical basis of the magnetic resonance technology, the EPR method is applicable only to the paramagnetic point defects on fused silica optical surfaces and so it cannot obtain the proportions and the densities of all the point defects. However, the EPR method could obtain the densities of the paramagnetic point defects on fused silica optical surfaces by comparing the paramagnetic point-defect densities on the detected specimen surfaces and that on the standardization specimen surfaces (the paramagnetic point-defect densities have been determined). It means that the densities of the point defects can be obtained as long as the point-defect proportions and the partial point-defect densities (paramagnetic point defects) can be obtained. Thus, it is significant to obtain the point-defect proportions to quantitatively explore the point defects on mechanically machined fused silica optical surfaces.

In this study, the point-defect types on mechanically machined fused silica optical surfaces were determined based on the obtained PL emission spectra on mechanically machined optical surfaces. Afterwards, the origins and evolution laws of various point defects were explored based on the PL properties and the properties (e.g., reaction rule and structural feature) of the point defects on mechanically machined optical surfaces. Then, the quantitative relationship among various point defects and the proportions of various point defects were obtained for the first time. Based on this, the point defect with the highest proportion was determined. This work quantitatively explored the proportions of the point defects on mechanically machined fused silica optical surfaces and systematically studied the evolution mechanisms among them. It is meaningful to fully reveal the comprehensive action mechanisms of various point defects on mechanically machined optical surfaces and provide new insights into elucidating the defect-induced laser damage mechanisms of mechanically machined fused silica optical components applied under intense laser irradiation from the atomic scale.

2. Experimental section

2.1 Sample preparation

The specimen employed is Corning 7980 fused silica glass, whose surfaces are polished to guarantee that the roughness meets the experimental requirements (Ra< 1 nm). The specimen shape is a rectangular parallelepiped with 20 mm length, 20 mm width and 4 mm height. This work focuses on exploring the evolution mechanism of the point defects on mechanically machined fused silica optical surfaces. However, the surface impurities on fused silica optical surfaces may interact with the intense laser irradiation, affecting the accuracy of the obtained experimental results in the PL-detection (see Section 2.3) and laser-induced damage (see Section 2.4) experiments. Thus, the surface impurities are eliminated with the removal of the Beilby layer by submerging fused silica specimens into HF-based etchant (5∼6% wt. HF +10∼12% wt. NH₄F, etching rate 10∼11 µm/h) for 1 min to remove ∼800 nm material from the surface to get rid of Ce metallic contaminants (due to the use of CeO₂ polishing slurry in the polishing processes of fused silica optical components) and obtain an ideal optical surface.

2.2 Defect preparation

To achieve the repeatability of the study on the evolution of the point defects, the dimensions and shapes of the surface defects must be reproducible. However, the shapes and dimensions of the machined induced surface defects in the mechanically machined processes are usually random, which poses a challenge to the reproducibility of the study. Thus, it is necessary to prepare some surface defects whose shapes and dimensions can be controlled. The Vickers indentations are prepared to model the surface defects on mechanically machined optical surfaces in many previously reported work [27–29]. Previous work indicates that most of the surface defects on mechanically machined fused silica optical surfaces are pitted surface defects [30]. Generally, pitted surface defects are introduced under the squeezing actions (normal pressure) of the grinding grains of the grinding heads. While the Vickers indentations could be prepared under the normal pressure of the Vickers hardness indenters [31]. Thus, the formation mechanisms of the Vickers indentations and the pitted surface defects are similar. Moreover, the dimensions of the Vickers indentations are fairly close to that of the pitted surface defects on mechanically machined fused silica optical surfaces [32]. To sum up, it is relatively reasonable to model the surface defects on mechanically machined fused silica optical surfaces by preparing Vickers indentations on fused silica optical surfaces. There are a large number of brittle regions in the Vickers indentations under large loads (5 N and 10 N) [27]. And the brittle regions located on fused silica optical surfaces would severely reflect the excitation laser in the PL-detection experiments [33,34]. In other words, it is difficult to accurately focus the excitation laser on the mechanically machined fused silica optical surfaces with the Vickers indentations under large loads (5 N and 10 N). Thus, the obtained PL emission spectra with the Vickers indentations under large loads (5 N and 10 N) may be less accurate than that with the Vickers indentations under small applied loads, because there are fewer brittle regions in the Vickers indentations under small applied loads. Therefore, the Vickers indentations under the applied loads of 1 N, 2 N and 3 N are chosen as examples to study. To explore the evolution of the point defects, various Vickers indentations have been prepared via the Vickers hardness tester (HRE-B) with different applied loads respectively. The retention time of the Vickers hardness tester is 10 ns. In order to explore the point defects located on mechanically machined fused silica optical surfaces, the optical surfaces with various indentations are tested in the PL-detection experiments and laser-induced damage experiments. To fully reveal the evolution of the point defects, it is necessary to study the point defects in different types of surface-defect zones. The variable-depth scratches could be formed under the shear pressure of the scratch tester (CSM Instruments SA). The scratches on the mechanically machined fused silica surfaces are also usually formed under the shear forces of the worn grinding grains of the grinding heads. Therefore, the variable-depth scratches prepared by the scratch tester and the scratches on the mechanically machined surfaces have similar formation mechanisms of plastic deformations and brittle fractures under the shear forces. Moreover, the dimensions of most scratches prepared by the scratch tester are all in the range of 1-100 µm, which is similar to that of the scratch-like surface defects on mechanically machined fused silica surfaces [20]. Thus, it is reasonable to choose the scratches to model the scratch-like surface defects. The length of the scratch is 500 µm. The applied load varies uniformly over time and ranges from 1-120 mN. The movement speed of the scratch awl is 20 µm/s in experiments. In order to explore the point defects in different surface defects, the optical surface with the scratch is tested in the PL-detection experiments. The main types of the surface defects on mechanically processed fused silica optical surfaces are pits, scratches and cracks [7,16,26]. While the main type of the subsurface defects is subsurface crack [7,32]. As mentioned above, the pits could be formed under the positive pressure of the grinding grains. The scratches are formed under the shear forces of the grinding grains. While the subsurface and surface cracks are all formed when the surface stress reaches the fracture energy of the material. In other words, the subsurface and surface cracks have the similar plastic deformation and brittle fracture mechanisms of the material under the mechanical forces. Besides, previous work indicates that the PL properties of the subsurface and surface defects are similar. In other words, the conclusions about the surface cracks obtained in this work could also provide a reference for the subsurface defects as mentioned above. To achieve the repeatability of the study on the evolution of the point defects, we have to make sure that the dimensions and shapes of the surface defects must be reproducible. Then, we need to prepare amounts of surface defects whose dimensions and shapes could be controlled to widely explore the characteristic information of the point defects located in the surface defects with different species and different dimensions to obtain some universal investigations and conclusions about these surface defects. Therefore, it may be more effective to use the method in this work as mentioned above to study the evolution of point defects on the subsurfaces/surfaces of fused silica optical components under the mechanical forces recently.

2.3 PL-detection experiment

PL-detection method is regarded as a promising method to explore the point defects on optical surfaces. The PL emission spectra are obtained in the PL-detection experiments. The optical pathway diagram is shown in Fig. 1. The wavelength of the excitation laser is adjustable in the range of 410 nm∼ 850 nm. And the wavelength of 440 nm is chosen because the PL intensity of the PL emission spectra on fused silica optical surfaces is relatively higher under the excitation laser irradiation with a wavelength of 440 nm in the PL-detection experiments in previous work. It has been widely accepted that the shape of the PL emission spectra has nothing to do with the wavelength of the excitation laser as long as the photon energy of the excitation laser is high enough [35–39]. Thus, the chosen excitation-laser wavelength may not affect the study on the evolution of the point defects. The excitation-laser pulse width is about 60 ps. The excitation laser beam has a spatial distribution of Gaussian profile on the detected fused silica optical surface with an effective spot diameter of 3 µm. The effective spot size of the excitation laser beam in the PL emission spectra experiment is measured by the instruments. To focus the excitation laser on the detected optical surface, a light source is employed to illuminate the whole detected optical surface. And a CCD is used to observe whether the excitation laser has focused on the detected optical surface. In this work, some representative positions (measure points) of the defect pits are detected in the PL-detection experiments. If the sizes of the desired measure points are small than the spot size of the excitation beam, the spot size of the excitation beam would not affect the measurement results. Otherwise, the spot size of the excitation beam would affect the measurement results. Due to the relatively small dimensions of these measure points, the spot size of the excitation beam could basically cover the measure points. It is also noteworthy that the spot size of the excitation beam could not be too large since we want to explore the PL properties of different representative positions of the defect pits. The detected optical surface is irradiated under the weak excitation laser shown in Fig. 1. As shown in Fig. 1, Reflector A is employed to focus the excitation laser on the detected optical surface and locate the detected surface-defect areas. When this step is finished, Reflector A would be withdrawn to let the released PL pass through the dichroic mirror and the filter shown in Fig. 1. The dichroic mirror shown in Fig. 1 is used to let the desired range of PL pass through, and the filter is used to avoid the action of the excitation laser. The transmission properties of the dichroic mirror and the filter are long-wave pass. The cutoff wavelengths of the dichroic mirror and the filter are 450 nm and 420 nm respectively. Finally, the PL emission spectra can be obtained by the spectrograph shown in Fig. 1. Multiple measurements were conducted at the same location in the PL-detection experiments to ensure that the PL emission spectrum of a measurement point is almost constant in the following five measurements to obtain the convincing PL-detection experimental results of the measurement point. Moreover, we specially prepared the surface defects whose dimensions and shapes could be controlled through the Vickers hardness indenters and the scratch tester. In other words, we can repeatedly obtain the surface defects with almost the same dimensions and shapes. To sum up, the experimental data in this work could be repeated.

Fig. 1. The optical path of the PL-detection experiment on mechanically machined fused silica optical surfaces.

Download Full Size | PDF

2.4 Laser-induced damage experiment

To explore the effect of the point defects on the laser damage resistance of the optical surface, the laser-induced damage experiments are conducted in this work. The light path of the laser-induced damage experiment is shown in Fig. 2. As shown in Fig. 2, an Nd: YAG laser generator is used for the laser initiation. The single-shot irradiation mode has a wavelength of 355 nm and a pulse width of 4 ns (FWHM). The output laser energy has a spatial distribution of Gaussian profile on the detected surface with an effective spot diameter of 380 µm, which is large enough to radiate the whole surface-defect area on the optical surface. The incident angle of the laser beam is 0°. Subsequently, the laser beam is focused on the detected surface by the optical lens. A He-Ne probe laser is also focused on the detected surface to assist in observing the laser-induced damage occurring on the detected surface. The laser-induced damage morphologies in the surface-defect zones are observed by a CCD camera after each laser pulse irradiation. R-on-1 test method refers to that the detected specimen surface is irradiated multiple times under the laser irradiation with a small linear increase of the laser flux at the same short interval until the laser damage occurs on the detected specimen surface. The R-on-1 test method was employed in the laser damage experiments.

Fig. 2. The optical path of the laser damage experiment on mechanically machined fused silica optical surfaces.

Download Full Size | PDF

2.5 Scanning electron microscopy and 3D microscope measurement

The surface morphologies of the prepared mechanically machined optical surfaces with various indentations are measured by scanning electron microscopy (HELIOS NanoLab 600i). The surface morphologies of the optical surface with the scratch and the machined defect-free optical surface are measured by the three-dimensional field microscope with a super depth (VHX-500).

3. Results and discussions

3.1 Determination of the point-defect types on mechanically machined optical surfaces

In this work, the characteristics of various point defects at different positions in different surface-defect zones are observed and analyzed in order to obtain the universal evolution laws (e.g., the origins, evolution laws and especially the quantitative relationship among various point defects) of the point defects on mechanically machined fused silica optical surfaces. Figure 3 shows the SEM images of the Vickers indentations under different loads (1 N, 2 N and 3 N as mentioned above). As shown in Fig. 3, the surface morphologies of the Vickers indentations under different applied loads (1N, 2N and 3N) are observed using the scanning electron microscope (SEM) to obtain some locations with representative structural features to study in this work. The Vickers-indentation zones can be roughly divided into several types of representative positions (the center of the indentation under the load of 1N. the center, edges and cracks of the indentation under the load of 2N and 3N) through observing the surface morphologies and PL properties (e.g., PL intensities) at different indentation positions (See Experimental Section). Thus, the indentation center is chosen to investigate the point defects located on the mechanically machined fused silica optical surface with the Vickers indentation under the load of 1 N shown in Fig. 3(a). As shown in Fig. 3(b), there are some brittle regions located on the edges of the Vickers indentation under the applied load of 2 N and there are some cracks located outside the Vickers indentation. Thus, the center, edge of the indentation and the crack outside the indentation are chosen as examples to study the point defects located on the mechanically machined fused silica optical surface with the Vickers indentation under the load of 2 N shown in Fig. 3(b). Similarly, the center, edge of the indentation and the crack outside the indentation are chosen to study the point defects located on the mechanically machined optical surface with the Vickers indentation under the load of 3 N shown in Fig. 3(c). It is noteworthy that the Vickers indentations whose applied loads are small than 1 N were not chosen to study in this work. The reasons are as follows. The excitation laser beam has a spatial distribution of Gaussian profile on the detected fused silica optical surface with an effective spot diameter of 3 µm. Figure 3(d) shows the SEM image of the Vickers indentation under the applied load of 0.5 N. As shown in Fig. 3(d), the difference between the spot diameter of the excitation laser beam and the dimension of the indentation (the applied load is small than 1 N) is small, which may affect the experimental results. For example, the excitation laser may be inevitably distributed on the other regions (e.g., the edge and the optical surface without surface defect) in the PL-detection experiment at the center of the Vickers indentation (the applied load is small than 1 N). Therefore, it may not be ideal to choose the Vickers indentations whose applied loads are small than 1 N to study. Besides, there are several regions with different brittleness-plasticity states and structural characteristics (e.g., plastic zone, fracture zone and the cracks with different species) on the Vickers indentations under the applied loads of 2 N and 3 N as shown in Fig. 3. The study on these regions with different species may obtain more meaningful conclusions and investigations. It is noteworthy that there is also a need for a plastic zone of the Vickers indentation without obvious brittle rupture of the material since it may be a kind of representative area with research value. As mentioned above, the excitation laser beam has an effective spot diameter of 3 µm. In order to obtain the PL properties of a plastic zone of the Vickers indentation without obvious brittle rupture of the material, we choose the center of the Vickers indentation under the applied load of 1 N to study since the difference between the spot diameter of the excitation laser beam and the dimension of the indentation (the applied load is small than 1 N) is small, which may affect the experimental results as mentioned above.

Fig. 3. SEM images of the Vickers indentations under different loads: (a) 0.5 N; (b) 1 N; (c) 2 N; (d) 3 N; (e) 5 N.

Download Full Size | PDF

Figure 3(e) shows the SEM image of the Vickers indentation under the applied load of 5 N. It can be seen from Fig. 3(e) that there are several broken areas around the edges of the Vickers indentation under the applied of 5 N. And these broken areas are usually accompanied by obvious surface irregularities (material collapses and warps), which would significantly affect the focus of the excitation laser beam and resultantly affect the accuracy of the experimental results in the PL-detection experiments. Thus, the Vickers indentations with the applied load of 5 N is not chosen to study in this work. As shown in Fig. 3, although there may be some brittle areas (e.g., cracks) in the Vickers indentations under the applied loads of 2N and 3N, the surface flatness (without obvious material collapses and warps) of the Vickers indentations is relatively good, which would not significantly affect the accuracy of the experimental results in the PL-detection experiments. It can be seen from Fig. 3 that there are several secondary cracks around the edges of the Vickers indentations under the applied loads of 2 N and 3 N. Therefore, the areas around the edges of the Vickers indentations under the applied loads of 2 N and 3 N could be regarded as the completely brittle zones, which are chosen to study in this work. However, there are almost no secondary cracks around the edge of the Vickers indentation under the applied load of 1 N. While the material warping at the edge of the Vickers indentation under the applied load of 1 N is due to the extrusion action of the edge of the Vickers indenter on the material. As mentioned above, the broken areas which are accompanied by obvious surface irregularities (material collapses and warps), which may affect the accuracy of the experimental results in the PL-detection experiments. Therefore, the area that has no obvious brittle characteristics other than the material warping may be not an ideal location for the further study in this work. In other words, the area around the edge of the Vickers indentation under the applied load of 1 N is not considered to be an ideal position to study (quantitatively study) in this work.

Due to the broad wavelength range of the PL released from the point defects on the mechanically machined fused silica optical surfaces, two spectrographs with different detection ranges (350-720 nm and 650-1100 nm) are both employed in the PL-detection experiments in this work. As shown in Fig. 4(a) and 4(b), the obtained PL emission spectra in the indentation zone under the load of 1 N consist of two parts. As shown in Fig. 4(a) and 4(b), seven fitted Gaussian components (four fitted Gaussian components shown in Fig. 4(a) and three fitted Gaussian components shown in Fig. 4(b) are obtained by analyzing the fitted PL emission spectra with Gaussian deconvolution [40]. It is noteworthy that the area of the fitting curves is equal to the sum of the area of the corresponding fitted Gaussian components with Gaussian deconvolution [40,41]. Previous work indicates that the point-defect types can be determined based on the obtained peak positions of the fitted Gaussian components [40]. Seven determined point defects are named in turn as Point defect A, Point defect B, Point defect C, Point defect D, Point defect E, Point defect F and Point G with the peak position increasing shown in Fig. 4(a) and 4(b). The peak positions of the seven types of point defects are Point defect A (2.54 eV), Point defect B (2.38 eV), Point defect C (2.18 eV), Point defect D (1.93 eV), Point defect E (1.62 eV), Point defect F (1.44 eV) and Point defect G (1.32 eV) shown in Fig. 4(a) and 4(b).

Fig. 4. (a) The PL emission spectrum (350-720 nm) in the indentation center (1 N-load). (b) The PL emission spectrum (650-1100 nm) in the indentation center (1 N-load).

Download Full Size | PDF

Figure 5(a) shows the peak positions (on the vertical axis) of all the point defects located in different positions (the centers, edges of the indentations and the cracks outside the indentations) on the mechanically machined fused silica optical surfaces with the Vickers indentations under different applied loads (1 N, 2 N and 3 N). I-VII shown in Fig. 5(a) are named in the same way as Fig. 3. It has been widely accepted that the peak position of a type of point defect is specific [40]. Therefore, Fig. 5(a) indicates that the types of the point defects in different positions (I-VII) on mechanically machined fused silica optical surfaces are the same. However, there are only 5-6 types of point defects but not 7 types of point defects in Position II, Position III, Position IV and Position VI, whose reason will be explained in detail later. The mean values of the peak positions of the same type of point defect (A-G) in different positions on the mechanically machined fused silica optical surfaces with the Vickers indentations under different applied loads are obtained shown in Fig. 5(a). Seven point defects on mechanically machined fused silica optical surfaces in this work are determined: Oxygen deficiency center II (ODCII), Self-trapped exciton (STE), E’-Center, Non-bridging oxygen hole center I (NBOHCI), Non-bridging oxygen hole center II (NBOHCII), Peroxy linkage (POL) and Silicon nanocluster point defects (with the peak position increasing). As shown in Fig. 5(b), the average value of the error rates between the mean values of the peak positions of the same type of point defect through the PL-detection experiments in this work and the corresponding previously reported point-defect peak positions is 2.92% and the maximum deviation between the mean values of the peak positions of the same type of point defect through the PL experiments and the previously reported point-defect peak positions is 0.1 eV [23]. Therefore, the test method used to explore the point defects on the mechanically machined fused silica optical surfaces in this work is effective. On basis of this, we can study the evolution of the point defects involved under the action of mechanical forces on mechanically machined fused silica surfaces.

Fig. 5. (a) Point-defect peak positions in different indentation positions on the mechanically machined fused silica optical surfaces with the Vickers indentation under different applied loads. (b) The comparison of the mean values of the point-defect peak positions through the PL-detection experiments in this work and the corresponding previously reported point-defect peak positions.

Download Full Size | PDF

The mechanically machined fused silica optical surfaces with various Vickers indentations under different applied loads (See Experimental Section) are tested under intense laser irradiation in the laser-induced damage experiments to study the effect of the point defects on the laser damage resistance of the mechanically machined optical surfaces. Figure 6 shows the surface morphologies of the Vickers indentations under different applied loads after the laser-induced damage. Figure 6 shows that the laser-induced damage usually occurs at the edges of the Vickers indentations under different applied loads.

Fig. 6. The surface morphologies of the Vickers indentations under different applied loads after the laser-induced damage: (a) 1 N; (b) 2 N; (c) 3 N.

Download Full Size | PDF

Therefore, the PL intensities at different locations in the Vickers-indentation zones especially the edges of the Vickers indentations under different applied loads (See Experimental Section) are tested in the PL-detection experiments in this work. I** shown in Fig. 7(a) refers to the edge of the Vickers indentation under the applied load of 1N. It can be concluded from Fig. 7 that there are usually high PL intensities around the edges of the Vickers indentations. Thus, it indicates that the PL intensity is closely related to the laser damage resistance. It is noteworthy that the PL intensities here refer to the peak (∼520 nm) in the PL emission spectrum (e.g., Fig. 4). The laser damage resistance of a position on a mechanically machined fused silica optical surface decreases with the PL intensity of the position. Moreover, the PL intensity is closely related to the point defects on the optical surfaces. Thus, it indicates that the point defects in the surface-defect zones on mechanically machined fused silica optical surfaces could severely threaten the laser damage resistance of the mechanically machined fused silica optical surfaces. The laser flux is continuously increased until the laser damage occurs in the indentation areas (R-on-1 method) in the laser damage experiments in this work. In other words, the laser flux when the laser damage occurs in the indentation area would meet the LIDT of the indentation area. The laser fluxes when the laser damage occurs in different indentation areas under different applied loads are different due to the different LIDTs of the different indentation areas. The LIDT of the indentation under the applied load of 3 N is smaller than that under the applied loads of 1 N and 2 N respectively. Therefore, the laser flux when the laser damage occurs in the indentation under the applied load of 3 N is also smaller than that under the applied loads of 1 N and 2 N. Thus, the damage is not induced in the center of indentation under 3N-load even though the PL intensity in the center of indentation under 3N-load is higher than that in the edge of indentation under 2N-load and 1N-load. Besides, for the same Vickers indentation, the laser damage usually occurs first where the PL intensity is highest among the whole indentation. It indicates that the point defects could significantly affect the LIDT of the surface defect.

Fig. 7. The PL intensities at different locations in the Vickers-indentation zones under different applied loads: (a) 1N; (2) 2N; (c) 3N.

Download Full Size | PDF

3.2 Relationship between the densities of oxygen-deficient and peroxide point defects

To explore the evolution of the point defects on mechanically machined fused silica optical surfaces, it is necessary to study the structural forms of these point defects. Figure 8 shows the structural forms of seven types of point defects on mechanically machined fused silica optical surfaces which have been obtained in Fig. 4. The purple balls represent Si atoms, the red balls represent O atoms and the black balls represent lone-pair electrons (unbonded electrons) [42]. The structures marked by dotted-line circles represent the intrinsic structures (unbroken structures) of fused silica. Previous work indicates that a large number of Si-O bonds would break in the surface-defect zones under the actions of the mechanical forces in the mechanical processing processes of the optical surfaces [43], which could introduce mass unbonded electrons (long-pair electrons). As shown in Fig. 8(a), there are two unbonded electrons in the ODCII point defect. There is an unbonded electron in both E’-Center and NBOHCI point defects shown in Fig. 8(c) and 8(d). As shown in Fig. 8(e), the interaction force between oxygens is so weak that the O-O bond is liable to break in the POL point defect under irradiation [44]. Thus, it can be assumed that there are two unbonded electrons in the POL point defect under laser irradiation shown in Fig. 8(e). And the POL point defect will be discussed in more detail later. There are two unbonded electrons in the NBOHCII point defect shown in Fig. 8(f). The Silicon nanoclusters (the determined point defect shown in Fig. 8(g) are induced by the introduction of impurities during the preparation processes of fused silica optical components [45]. The distribution of the Silicon nanoclusters has a certain randomness. Thus, it can explain the reason why there are only 5-6 types of point defects but not 7 types of point defects in some positions (the lack of the Silicon nanoclusters). The introduction of Silicon nanoclusters on mechanically machined fused silica optical surfaces has nothing to do with the mechanical forces during the machining processes of the mechanically machined optical surfaces and the surface defects (the main aim of this work) induced by the mechanical forces [45]. Herein, Silicon nanoclusters would not be studied in detail in this work. STE point defects (the determined point defect shown in Fig. 8(b) are formed under mechanical forces before Si-O bonds are completely broken [46]. One of the purposes of this work is to explore the effect of the Si-O bond fracture (will be discussed in detail later) on the relationship among the densities of various point defects on mechanically machined optical surfaces. Thus, STE point defects would not be studied in detail in this work, whose reason will also be explained in detail later in this work.

Fig. 8. Structural forms of seven types of point defects on mechanically machined optical surfaces.

Download Full Size | PDF

According to the previously reported work, the transformation mechanisms of five types of point defects on mechanically machined optical surfaces are shown as follows [46,47].

The E’-Center point defects and NBOHCI point defects could be formed when Si-O bonds break under the actions of the mechanical forces in the machining processes of the mechanically machined optical surfaces shown in Formula (1):

(1)$$\equiv \textrm{Si-O-Si} \equiv \textrm{ } \to \textrm{ } \equiv \textrm{Si}\cdot \textrm{ + }\cdot \textrm{O-Si} \equiv$$

POL point defects would be formed by the polymerization of two NBOHCI point defects shown in Formula (2):

(2)$$\equiv \textrm{Si-O}\cdot \textrm{ + }\cdot \textrm{O-Si} \equiv \textrm{ } \to \textrm{ } \equiv \textrm{Si-O-O-Si} \equiv$$

However, POL point defects are very unstable and so they tend to be decomposed into NBOHCI point defects and NBOHCII point defects under irradiation shown in Formula (3).

(3)$$\equiv \textrm{Si-O-O-Si} \equiv \textrm{ } \to \textrm{ } \equiv \textrm{Si-O}\cdot \textrm{ + :O-Si} \equiv$$

ODCII point defects could be further formed based on E’-Center point defects under mechanical forces in the machining processes of mechanically machined optical surfaces shown in Formula (4):

(4)$$\equiv \textrm{Si}\cdot \textrm{ } \to \textrm{ = Si}: \textrm{ + }\cdot \textrm{O-Si} \equiv$$

As shown in Formula (1)–(4), the types of the point defects located on the fused silica surfaces and their conversion mechanisms between different kinds of point defects have been reported in many previous investigations. Therefore, the types of the point defects located on the fused silica surfaces and their conversion mechanisms (Between two kinds of point defects) have been widely acknowledged. However, the evolution of all the point defects involved under the action of mechanical forces on mechanically machined fused silica surfaces has not been systematically and quantitatively studied till now, which poses a challenge to addressing the issues of the laser-induced damage on the mechanically processed optical surfaces under intense laser irradiation. Therefore, this work systematically explores the origins, evolution laws and the quantitative relationship among point defects under the action of mechanical forces in order to address these issues.

To sum up, the transformation mechanisms among the point defects located on mechanically machined optical surfaces in the mechanical processes of the fused silica optical components can be summarized shown in Fig. 9. Firstly, Si-O bonds in the intrinsic structures of fused silica on mechanically machined optical surfaces would break, inducing an equal number of E’-Center point defects and NBOHCI point defects. And an equal number of ODCII point defects and NBOHCI point defects could be further formed by E’-Center point defects under mechanical forces. POL point defects could be formed by the polymerization of two NBOHCI point defects. However, POL point defects would be decomposed into an equal number of NBOHCI and NBOHCII point defects.

Fig. 9. Transformation mechanisms among the point defects located on mechanically machined fused silica optical surfaces.

Download Full Size | PDF

Herein, based on the transformation mechanisms among the point defects under the action of mechanical forces, the point defects on mechanically machined fused silica optical surfaces are further explored. Figure 9 shows the transformation mechanisms among the point defects located on mechanically machined fused silica optical surfaces. As shown in Fig. 9, E’-Center point defects and NBOHCI point defects can be regarded as ‘basic point defects’. And the rest point defects (NBOCHI point defect, NBOHCII point defect and POL point defect) are derived from the basic point defects (E’-Center point defect and NBOHCI point defect). Moreover, the transformation laws between the reactants (reaction point defects) and the products (birth point defects) could be described quantitatively in each reaction process shown in Formula (1)–(4). In theory, the relationship between point-defect densities on mechanically machined fused silica optical surfaces could be obtained. As shown in Fig. 9, in this work, we divided the point defects on mechanically machined optical surfaces into oxygen-deficient point defects (Part A) and peroxide point defects (Part B). Part A consists of the E’-Center point defect and the ODCII point defect. Part B consists of the NBOHCI point defect, the NBOHCII point defect and the POL point defect. Herein, efforts have been paid to explore the quantitative relationship between the densities of Part A and Part B point defects.

As shown in Fig. 9 and Formula (1), E’-Center point defects and NBOHCI point defects generate in equal quantities as Si-O bonds break. And the generation of an ODCII point defect is always accompanied by the generations of two NBOHCI point defects shown in Fig. 9, Formula (1) and Formula (4). Thus, if the NBOHCI point defect would not react and generate other point defects (POL point defect and NBOHCII point defect), the number of NBOHCI point defects N_NI in theory should be described as below:

(5)$$N_{NI}\textrm{ = }N_{E} + 2N_{O}$$

where N_E is the number of E’-Center point defect, N_o is the number of ODCII point defect. However, NBOHCI point defect would react and generate other point defects (POL point defect and NBOHCII point defect) as shown in Fig. 9 and Formula (2). Thus, it can explain the reason why there are only 5-6 types of point defects but not 7 types of point defects in some positions (The lack of the NBOHCI point defect at some positions). POL point defect would be formed by the polymerization of two NBOHCI point defects shown in Fig. 9 and Formula (2). As shown in Fig. 9 and Formula (3), a POL point defect could be decomposed into an NBOHCI point defect and an NBOHCII point defect in equal quantities under irradiation. To sum up, a POL point defect can be equivalent to two NBOHCI point defects while an NBOHCII point defect can be equivalent to an NBOHCI point defect. The number of the POL point defect is recorded as N_P while the number of the NBOHCII point defect is recorded as N_NII. Therefore, the relationship between the numbers of the point defects on mechanically machined optical surfaces should be described below:

(6)$$N_E + 2N_O = N_{NI} + N_{NII} + 2N_P$$

Therefore, the relationship between the densities of the point defects on mechanically machined optical surfaces should be described as below:

(7)$$n_E + 2n_O = n_{NI} + n_{NII} + 2n_{P}$$

where n_E is the density of the E’-Center point defect, n_o is the density of the ODCII point defect. n_NI is the density of the NBOHCI point defect, n_NII is the density of the NBOHCII point defect, n_P is the density of the POL point defect. To sum up, it can be concluded from the above theoretical analysis that there is a definite quantitative relationship (see Formula (7)) between the densities of the oxygen-deficient point defects and that of the peroxide point defects on mechanically machined optical surfaces based on the rupture of Si-O bonds which can also explain the reason why STE point defects would not be studied in detail in this work.

3.3 Relationship between PL, the densities of the unbonded electrons and the point defects

Ground-state electrons at the Valence band on mechanically machined optical surfaces could transit to the Conduction band and transform into free electrons through absorbing photon energy under irradiation [25,48]. And the mechanically machined optical surfaces would suffer laser-induced damage when the free-electron densities on mechanically machined optical surfaces reach the critical free-electron density [49,50]. It has been widely accepted that the ground-state electrons on the mechanically machined optical surfaces with surface defects are more easily to be ionized to transform into free electrons with the assistance of various point defects located on mechanically machined optical surfaces. It is also the reason why point defects could change the optical absorption properties of the mechanically machined optical surfaces and sharply reduce their laser damage resistance. However, locating the ground-state electrons on mechanically machined optical surfaces which tend to be ionized is still an unsolved issue. Since the unbonded electrons in point defects are prone to react (with powerful chemical activity) shown in Formula (1)–(4), the unbonded electrons are assumed to be the ground-state electrons on mechanically machined optical surfaces which tend to be ionized to transform into free electrons under irradiation (with powerful optical activity). Moreover, the unbonded electrons are addressed as ‘active ground-state electrons under irradiation’ (AGE) with powerful optical activity while the other electrons in various point defects and the intrinsic structures of fused silica are addressed as ‘stable ground-state electrons under irradiation’ (SGE) with poor optical activity in this work.

It has been well known that PL would be released when the ionized free electrons in the point defects located on mechanically machined fused silica optical surfaces undergo electron attenuation [51]. Therefore, the PL emission spectra obtained in the PL-detection experiments could effectively characterize the electron transition processes in the point defects located on mechanically machined fused silica optical surfaces. Previous work shows that the PL released is proportional to the electron density at the higher energy level in the electron-transition processes on mechanically machined optical surfaces [52,53]. Some work also shows that the electron density at the higher energy level is roughly proportional to that at the Valence band in the electron-transition processes on mechanically machined optical surfaces under weak irradiation [18,54]. Thus, PL can be assumed to be roughly proportional to the ground-state electron (AGE) density (AGE is thought to can induce PL) at the Valence band in the electron-transition processes on mechanically machined optical surfaces. Thus, the PL released in the electron transition processes in the point defects on mechanically machined optical surfaces can be characterized by the obtained PL emission spectra in the PL-detection experiments in this work. Previous work [25,55] shows that the obtained Gaussian components fitted in the obtained PL emission spectra correspond to different point defects on mechanically machined optical surfaces shown in Fig. 4. In other words, each type of point defect has its corresponding Gaussian component in the obtained PL emission spectra shown in Fig. 4. The detected PL released by each type of point defect in the PL emission spectra can be represented by the envelope area of the corresponding Gaussian component curve in the PL emission spectra shown in Fig. 4. For example, the detected PL released by the ODCII point defect can be represented by the envelope area of S1 shown in Fig. 4(a). To sum up, it is reasonable that the envelope area of the corresponding Gaussian component curve of a type of point defect is assumed to be proportional to the ground-state electron density of this type of point defect. It is noteworthy that the ground-state electrons which tend to be ionized in the surface-defect zones on mechanically machined optical surfaces under irradiation are assumed to be AGEs mentioned above. According to the obtained conclusion that the unbonded electrons (AGEs) in the point defects tend to transform into free electrons through absorbing photon energies as mentioned above, the AGE density n_ei of a type of point defect is related to the density of this type of point defect (n_i) on mechanically machined optical surface and the number of AGE in a point defect (m), which can be described as below:

(8)$$n_{ei} = m \ast n_i$$

As shown in Fig. 9, there are several AGEs in various point defects. Thus, m for different point defects can be determined. As shown in Fig. 9, m for ODCII, E’-Center, NBOHCI, NBOHCII and POL point defects are 2, 1, 1, 2, 2, respectively. Thus, ODCII, POL and NBOHCII point defects are considered to have a greater impact on the laser damage resistance of mechanically machined optical surfaces because they can provide more AGEs to induce laser damage. It has been widely accepted that the PL intensity induced by the substance (could induce PL under irradiation) under weak irradiation is approximately proportional to the density of the substance. Herein, the corresponding PL intensity of each type of point defect is firstly obtained in this work. It has been widely accepted that most of the point defects would form defect clusters rather than in the form of isolated point defects. Several sub-bandgap defect levels would be introduced on the fused silica surfaces under the actions of the defect clusters. And the sub-bandgap defect levels could significantly promote multiphoton ionization and sharply reduce the LIDTs of the fused silica surfaces. The sub-bandgap defect levels are determined for a specific location on the fused silica surfaces and the electrons in various point defects at the specific location would absorb the photon energy and transit to the Conduction band (transform into free electrons) with the assistance of the same sub-bandgap defect levels. In other words, the electrons in various point defects of the defect clusters could be roughly considered as a class of electrons that have the same difficulty in making transitions and have the same ability to absorb photons. The excited electrons from the various point defects would all decay into ground state electrons and induce PL. However, the wavelength of the PL induced in various point defects are different due to the different chemical structure forms of the various point defects. To sum up, the density of the substance (the electrons could induce PL under irradiation) in each type of point defect is approximately proportional to the corresponding induced PL intensity in the PL emission spectra. Therefore, the relative proportion of the density of the substance (the electrons could induce PL under irradiation) in each type of point defect can be obtained. Moreover, the substances (the electrons could induce PL under irradiation) which could induce PL under irradiation are determined in this work. This work found that the lone-pair electrons (unbonded electrons) of the point defects can induce PL under excitation irradiation. Then, the number of lone-pair electrons of the point defects with different species is also determined. On basis of this, the relative proportion of the density of each type of point defect can be obtained.

3.4 Experiment validation

As mentioned above, the envelope area of the corresponding Gaussian component curve of a type of point defect is assumed to be proportional to the AGE density of this type of point defect. Thus, the relationship between the envelope areas of the corresponding Gaussian component curves of various point defects on the mechanically machined fused silica optical surfaces and the corresponding AGE densities of the point defects are shown in Formula (9).

(9)$$\left\{ \begin{array}{l} S1 = kn_{eO}\\ S2 = kn_{eE}\\ S3 = kne_{NI}\\ S4 = kne_{NII}\\ S5 = kn_{eP} \end{array} \right.$$

where the envelope areas of the corresponding Gaussian component curves of ODCII, E’-Center, NBOHCI, NBOHCII and POL point defects are named as S1, S2, S3, S4, S5, respectively shown in Fig. 4(a) and 4(b). k is the ratio between the envelope area of the corresponding Gaussian component curve of a type of point defect and the AGE density of this type of point defect on mechanically machined optical surface. n_eO, n_eE, n_eNI, n_eNII and n_eP are the AGE densities of ODCII, E’-Center, NBOHCI, NBOHCII and POL point defects on mechanically machined optical surfaces. The AGE density n_e of a type of point defect is related to the density of this type of point defect (n_i) on the mechanically machined optical surface and the number (m) of AGEs in a point defect on the mechanically machined optical surface shown in Formula (8). Therefore, based on the obtained relationship between the densities of the point defects on the mechanically machined optical surfaces shown in Formula (7), the point-defect densities of Part A and Part B can be described as below:

(10)$$\begin{array}{l} D_1 = 2 \times ( \frac{{S1}}{{k \times 2}}) + 1 \times ( \frac{{S2}}{k}) \textrm{ = }\frac{{S1}}{k} + \frac{{S2}}{k}\\ \end{array}$$

(11)$$D_2 = 1 \times ( \frac{{S3}}{k}) + 1 \times ( \frac{{S4}}{{2 \times k}}) + 2 \times ( \frac{{S5}}{{2 \times k}}) = \frac{{S3}}{k} + \frac{{S4}}{{2 \times k}} + \frac{{S5}}{k}$$

where D₁ represents the point-defect densities of Part A (ODCII point defect and E’-Center point defect), D₂ represents the point-defect densities of Part B (NBOHCI point defect, NBOHCII point defect and POL point defect). According to Formula (7), (10) and (11), the relationship between the point-defect densities of Part A and Part B can be described as below:

(12)$$\frac{{S1}}{k} + \frac{{S2}}{k} = \frac{{S3}}{k} + \frac{{S4}}{{2 \times k}} + \frac{{S5}}{k}$$

Thus, the relationship between the envelope areas of the corresponding Gaussian component curves of various point defects should be described as below:

(13)$$S1 + S2 = S3 + \frac{{S4}}{2} + S5$$

(14) $$E_A = S1 + S2$$

(15)$$E_B = S3 + \frac{{S4}}{2} + S5$$

where E_A and E_B represent the envelope areas of the corresponding Gaussian component curves of Part A and Part B, respectively. Figure 10 shows the envelope areas of the corresponding Gaussian component curves of Part A and Part B (E_A and E_B) obtained in the PL-detection experiments. The names of the different positions shown in Fig. 10 refer to Fig. 3. As shown in Fig. 10, E_A and E_B vary from place to place in the indentation zones. However, E_A is about equal to E_B in each position. The average deviation rate shown in Fig. 10 refers to the average value of the deviation rates between E_A and E_B in different positions (4.64%). It is noteworthy that the two detection ranges (350-720 nm and 650-1100 nm) overlap in the PL emission spectra shown in Fig. 4 may make a partial contribution to the deviation between E_A and E_B. It indicates that E_A can be considered to be equal to E_B in the PL-detection experiments. And the maximum deviation between the envelope areas of the corresponding Gaussian component curves of Part A (oxygen-deficient point defects) and Part B (peroxide point defects) is 5.5 × 10³. It verifies the conclusions that the unbonded electrons in the point defects tend to transform into free electrons through absorbing photon energies (see Formula (8)) and there is a definite quantitative relationship (see Formula (7)) between the densities of the oxygen-deficient point defects and that of the peroxide point defects on mechanically machined optical surfaces.

Fig. 10. The comparison of the envelope areas of the corresponding Gaussian component curves of Part A (oxygen-deficient point defects) and Part B (peroxide point defects) tested in the PL-detection experiments.

Download Full Size | PDF

To further verify the conclusions that the unbonded electrons with powerful chemical activities in the point defects tend to transform into free electrons through absorbing photon energies and there is a definite quantitative relationship between the densities of the oxygen-deficient point defects and that of the peroxide point defects on mechanically machined optical surfaces, more types of optical surfaces with other surface defects are chosen to study in this work shown in Fig. 11. As shown in Fig. 11(a) and 11(b), the surface morphologies of the optical surface with the scratch and the machined defect-free optical surface are observed using the VHX-500 three-dimensional microscope (See Experimental Section). As shown in Fig. 11(a), the scratch is prepared via a scratch tester (CSM Instruments SA). The length of the scratch shown in Fig. 11(a) is 500 µm. To obtain the variable-depth scratch (See Experimental Section), the applied load varies uniformly with the sliding distance of the awl (the sliding speed of the awl is constant) and ranges from 1-120 mN. The movement speed of the scratch awl is 20 µm/s. To further verify the above conclusions obtained in this work, the defect-free position (surface defect is invisible) on the mechanically machined fused silica optical surface is explored shown in Fig. 11(b). As shown in Fig. 11(a), four equally spaced (160 µm) test points of the scratch are chosen to study in this work, which are named as I*, II*, III* and IV* in turn. Similarly, two test points are chosen to study in this work and there are no surface defects around the test points under the observation of the CCD in the PL-detection experiments. The test points are named as V* and VI*, respectively shown in Fig. 11(b).

Fig. 11. (a) The surface topography of the optical surface with the scratch imaged by VHX-500 three-dimensional microscope. (b) The surface topography of the machined defect-free optical surface imaged by VHX-500 three-dimensional microscope. (c) The comparison of the envelope areas of the corresponding Gaussian component curves of Part A (oxygen-deficient point defects) and Part B (peroxide point defects) on the optical surface with the scratch and the machined defect-free optical surface.

Download Full Size | PDF

Figure 11 shows the envelope areas of the corresponding Gaussian component curves of Part A and Part B (E_A and E_B) at test points I*- VI* as mentioned above in the PL-detection experiments. As shown in Fig. 11(c), E_A and E_B at different test points (I*- VI*) are different. Nonetheless, E_A is roughly equal to E_B at each test point. The average deviation rate of different test positions is 9.8% shown in Fig. 11(c), which indicates that E_A can be considered equal to E_B in the PL-detection experiments. It is also noteworthy that the two detection ranges (350-720 nm and 650-1100 nm) overlap in the PL emission spectra shown in Fig. 4 may make a partial contribution to the deviation between E_A and E_B. And the maximum deviation between the envelope areas of the corresponding Gaussian component curves of Part A (oxygen-deficient point defects) and Part B (peroxide point defects) is 6.5 × 10³. Therefore, the obtained conclusions that the unbonded electrons in the point defects tend to transform into free electrons through absorbing photon energies and there is a definite quantitative relationship between the densities of the oxygen-deficient point defects and that of the peroxide point defects on mechanically machined optical surfaces shown in Formula (7) and (8) are further verified. Based on Formula (14) and (15), the proportion of each point defect P_i can be obtained.

(16)$${P_i} = \frac{{{S_i}/{m_i}}}{{\sum\limits_{j = 1}^5 {( {S_i}/{m_i}) } }}$$

where S_i is the envelope area of the corresponding Gaussian component curve of this type of point defect. ${m_i}$ is the number of the AGEs in a point defect. $\sum\limits_{j = 1}^5 {{S_j}/{m_j}}$ means that the proportion of a point defect in this work refers to the proportion of this point defect in the five point defects. (ODCII, E’-Center, NBOHCI, NBOHCII and POL point defects). Figure 12 shows the average proportions of the point defects (Part A and Part B) in different positions (I-VII and I*-VI*). As shown in Fig. 12, E’-Center point defect accounts for the highest proportion (45.45%) while ODCII point defect accounts for the lowest proportion (4.73%) among the point defects on mechanically machined optical surfaces.

Fig. 12. The average proportions of the point defects in different positions (I-VII and I*-VI*) on mechanically machined fused silica optical surfaces.

Download Full Size | PDF

For Part A (oxygen-deficient point defects), there is a large gap between the proportions of the E’-Center point defect (45.45%) and ODCII point defect (4.73%). It may be related to the brittle-plastic states of the surface defects. For Part B (peroxide point defects), the difference between the average proportions of the NBOHCI (11.27%), NBOHCII (18.76%) and POL (19.79%) point defects is small. It may be related to the evolution mechanism among these three types of point defects. In summary, the proportions of the point defects on mechanically machined optical surfaces are obtained in this work, which is of great importance to quantitatively characterize the effect of the point defects on the laser damage resistance of the mechanically machined fused silica optical surfaces.

The surface defects with different species have different plastic deformation and brittle fracture mechanisms. The characteristic information (e.g., type, proportion and amount) of the point defects with different species and different dimensions were widely explored to study the evolution of the point defects involved under the action of mechanical forces. Based on the study, some universal investigations and conclusions about these surface defects have been obtained in this work. Therefore, the obtained investigations and conclusions in this work could be applicable to almost all the surface defects on mechanically processed fused silica optical surfaces in industrial production. Moreover, the study on the evolution of the point defects located on the optical surfaces of the optical components (e.g., KDP, fused silica, BK7 glasses and K9 glasses) was rarely reported till now. However, it is very beneficial to the improvement of the laser damage resistance of the optical components. Therefore, this work could also provide a reference for the study on the point defects located on the mechanically processed optical surfaces of other types of optical components. Besides, the proportions of the point defects are obtained for the first time through the study on the PL properties of various surface defects. On basis of this, researchers could explore the formation mechanism of the sub-bandgap defect energy levels fundamentally. It is also noteworthy that the defect-induced laser damage severely threatens the laser damage resistance of the fused silica optical components applied in many high-power laser systems. The lack of a definite relationship between the feature information of the surface defects and the LIDTs of the mechanically machined optical surfaces poses a challenge to the characterization for the influence of the surface defects and the revelation of the defect-induced laser damage mechanisms [15,16,33]. In this work, the proportions of the point defects on mechanically machined fused silica optical surfaces are obtained for the first time. On basis of this, the relationship between the feature information of the surface-defect zones and the LIDTs of the mechanically machined optical surfaces could be systematically explored. The LIDTs of the mechanically machined optical surfaces could be further characterized through the nondestructive PL-detection experiments.

4. Conclusion

Due to the distinct effects of various point defects on the laser damage resistance of the optical surfaces, it is necessary to explore the comprehensive effect of various point defects. Thus, the proportions and the evolution of the point defects are explored in this work. The following conclusions can be obtained:

(1) Seven types of point defects (ODCII, STE, E’-Center, NBOHCI, NBOHCII, POL and Silica nanocluster) are determined by analyzing the steady PL emission spectra with Gaussian deconvolution. The rest point defects (ODCII, NBOHCII and POL) are derived from the two basic point defects (E’-Center and NBOHCI).
(2) Based on the structural features, reaction rules of the determined point defects and the electron transition theory as well as the experimental results, it is concluded that the unbonded electrons in various point defects could provide the ground-state electrons which tend to transform into free electrons under intense irradiation.
(3) A definite quantitative relationship between the densities of the oxygen-deficient point defects (ODCII and E’-Center) and that of the peroxide point defects (NBOHCI, NBOHCII and POL) on mechanically machined fused silica optical surfaces is found in this work.
(4) The specific relationship between the PL and the proportion of the point defect on mechanically processed fused silica optical surface is constructed for the first time. On basis of this, the proportions of the point defects with different species on mechanically processed fused silica optical surfaces are obtained for the first time through the study on the PL properties of various surface defects with different dimensions and shapes.
(5) E’-Center point defect is found to have the largest proportion. The ODCII, POL and NBOHCII point defects are considered to have greater impacts on the laser damage resistance of the mechanically machined optical surfaces since there are more AGEs in these point defects.

To sum up, this work quantitatively explores the proportions of the point defects and systematically explores the evolution of the point defects on mechanically machined optical surfaces. This work is beneficial for fully revealing the comprehensive action mechanisms of various point defects and providing new insights in elucidating the defect-induced laser damage mechanisms on mechanically machined fused silica optical components under intense laser irradiation from the atomic scale. Furthermore, this work is very beneficial to the improvement of the LIDTs of the optical components applied in high-power laser systems.

Funding

National Natural Science Foundation of China (52175389, 52235010, 52293403); Consolidation Program for Fundamental Research (2019JCJQZDXX00); Young Elite Scientists Sponsorship Program by CAST (2018QNRC001); China Postdoctoral Science Foundation (2018T110288); Natural Science Foundation of Heilongjiang Province (YQ2021E021); Self-Planned Task of State Key Laboratory of Robotics and System (HIT) (SKLRS201718A, SKLRS201803B).

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. P. A. Baisden, L. J. Atherton, R. A. Hawley, T. A. Land, J. A. Menapace, P. E. Miller, M. J. Runkel, M. L. Spaeth, C. J. Stolz, T. I. Suratwala, P. J. Wegner, and L. L. Wong, “Large Optics for the National Ignition Facility,” Fusion Sci. Technol. 69(1), 295–351 (2016). [CrossRef]

2. Y. C. Dong, S. Chertopalov, K. Maleski, B. Anasori, L. Y. Hu, S. Bhattacharya, A. M. Rao, Y. Gogotsi, V. N. Mochalin, and R. Podila, “Saturable Absorption in 2D Ti3C2 MXene Thin Films for Passive Photonic Diodes,” Adv. Mater. 30(10), 1705714 (2018). [CrossRef]

3. G. Abromavicius, T. Juodagalvis, R. Buzelis, K. Juskevicius, R. Drazdys, and S. Kicas, “Oxygen plasma etching of fused silica substrates for high power laser optics,” Appl. Surf. Sci. 453, 477–481 (2018). [CrossRef]

4. N. Hou, L. C. Zhang, Y. Zhang, and F. H. Zhang, “On the Ultra-Precision Fabrication of Damage-Free Optical KDP Components: Mechanisms and Problems,” Crit. Rev. Solid State Mat. Sci. 44(4), 283–297 (2019). [CrossRef]

5. T. I. Suratwala, W. A. Steele, M. D. Feit, K. Moreno, M. Stadermann, J. Fair, K. Chen, A. Nikroo, K. Youngblood, and K. Wu, “Polishing and local planarization of plastic spherical capsules using tumble finishing,” Appl. Surf. Sci. 261, 679–689 (2012). [CrossRef]

6. Y. J. Chai, M. P. Zhu, H. B. Xing, H. Wang, Y. Cui, and J. D. Shao, “Multilayer deformation planarization by substrate pit suturing,” Opt. Lett. 41(15), 3403–3406 (2016). [CrossRef]

7. J. Cheng, H. Yang, Q. Liu, L. J. Zhao, Z. C. Liu, H. N. Liu, T. Z. Wang, Y. Xiao, K. H. Hu, M. J. Chen, and J. B. Tan, “Characterization of manufacturing-induced surface scratches and their effect on laser damage resistance performance of diamond fly-cut KDP crystal,” Results Phys. 15, 102753 (2019). [CrossRef]

8. Q. Y. Feng, H. X. Deng, S. Z. Huang, B. Li, X. Xiang, L. Li, B. Y. Wang, W. G. Zheng, X. D. Yuan, S. A. Li, H. D. Yang, and X. T. Zu, “Strong UV laser absorption source near 355 nm in fused silica and its origination,” Opt. Express 29(20), 31849–31858 (2021). [CrossRef]

9. Y. Esqueda-Barron, M. Herrera, and S. Camacho-Lopez, “ZnO synthesized in air by fs laser irradiation on metallic Zn thin films,” Appl. Surf. Sci. 439, 681–688 (2018). [CrossRef]

10. L. X. Sun, H. J. Liu, J. Huang, X. Ye, H. D. Xia, Q. Z. Li, X. D. Jiang, W. D. Wu, L. M. Yang, and W. G. Zheng, “Reaction ion etching process for improving laser damage resistance of fused silica optical surface,” Opt. Express 24(1), 199–211 (2016). [CrossRef]

11. Y. Li, H. W. Yan, K. Yang, C. Z. Yao, Z. Q. Wang, X. S. Zou, C. Y. Yan, X. D. Yuan, X. Ju, and L. M. Yang, “Surface molecular structure defects and laser-induced damage threshold of fused silica during a manufacturing process,” Sci. Rep. 7(1), 17870 (2017). [CrossRef]

12. L. X. Sun, T. Shao, X. D. Zhou, W. H. Li, F. F. Li, X. Ye, J. Huang, S. F. Chen, B. Li, L. M. Yang, and W. G. Zheng, “Understanding the effect of HF-based wet shallow etching on optical performance of reactive-ion-etched fused silica optics,” RSC Adv. 11(47), 29323–29332 (2021). [CrossRef]

13. Q. Chen, H. Wang, R. C. Dai, Z. P. Wang, X. P. Tao, W. Zhao, and Z. M. Zhang, “Correlation between photoluminescence properties of surface defects and laser-induced damage threshold of fused silica,” Laser Part. Beams 2021, 5530410 (2021). [CrossRef]

14. B. Li, C. Y. Hou, C. X. Tian, J. L. Guo, X. Xiang, X. L. Jiang, H. J. Wang, W. Liao, X. D. Yuan, X. D. Jiang, and X. T. Zu, “Layer by layer exposure of subsurface defects and laser-induced damage mechanism of fused silica,” Appl. Surf. Sci. 508, 145186 (2020). [CrossRef]

15. N. Shen, M. J. Matthews, S. Elhadj, P. E. Miller, A. J. Nelson, and J. Hamilton, “Correlating optical damage threshold with intrinsic defect populations in fused silica as a function of heat treatment temperature,” J. Phys. D: Appl. Phys. 46(16), 165305 (2013). [CrossRef]

16. L. J. Zhao, J. Cheng, M. J. Chen, X. D. Yuan, W. Liao, H. Yang, Q. Liu, and H. J. Wang, “Development of Integrated Multi-Station System to Precisely Detect and Mitigate Surface Damage on Fused Silica Optics,” Int. J. Precis. Eng. Man. 22(1), 95–105 (2021). [CrossRef]

17. C. W. Carr, H. B. Radousky, and S. G. Demos, “Wavelength dependence of laser-induced damage: Determining the damage initiation mechanisms,” Phys. Rev. Lett. 91(12), 127402 (2003). [CrossRef]

18. G. Duchateau, M. D. Feit, and S. G. Demos, “Strong nonlinear growth of energy coupling during laser irradiation of transparent dielectrics and its significance for laser induced damage,” J. Appl. Phys. 111(9), 093106 (2012). [CrossRef]

19. N. Ollier, K. Piven, C. Martinet, T. Billotte, V. Martinez, D. R. Neuville, and M. Lancry, “Impact of glass density on the green emission and NBOHC formation in silica glass: A combined high pressure and 2.5 MeV electron irradiation,” J. Non. Cryst. Solids 476, 81–86 (2017). [CrossRef]

20. F. Geng, H. P. Cheng, Q. H. Zhang, M. C. Liu, and Y. G. Li, “Ultraviolet laser damage properties of single-layer SiO2 film grown by atomic layer deposition,” Opt. Mater. Express 10(8), 1981–1990 (2020). [CrossRef]

21. L. Skuja, “Direct singlet-to-triplet optical absorption and luminescence excitation band of the twofold-coordinated silicon center in oxygen-deficient glassy SiO₂,” J. Non-Cryst. Solids 167(3), 229–238 (1994). [CrossRef]

22. L. Skuja, “Optically active oxygen-deficiency-related centers in amorphous silicon dioxide,” J. Non-Cryst. Solids 239(1-3), 16–48 (1998). [CrossRef]

23. J. Fournier, P. Grua, J. Neauport, E. Fargin, V. Jubera, D. Talaga, A. Del Guerzo, G. Raffy, and S. Jouannigot, “Temperature dependence of luminescence for different surface flaws in high purity silica glass,” Opt. Mater. Express 3(1), 1–10 (2013). [CrossRef]

24. O. Gurbuz, S. Guner, O. Buyukbakkal, and S. Caliskan, “Structural, optical, and conducting properties of crystalline ZnO:Co thin films grown by reactive electron beam deposition,” J. Magn. Magn. Mater. 373, 90–95 (2015). [CrossRef]

25. J. Qian, G. D. Wang, D. Y. Shen, K. Y. Lou, Q. Fu, Z. Y. Xia, and Q. Z. Zhao, “Tunable point defects in hydroxyl fused silica enabled by ultrashort laser pulses: photostimulated luminescence and functional module fabrication,” Opt. Mater. Express 10(5), 1241–1248 (2020). [CrossRef]

26. J. Cheng, M. J. Chen, W. Liao, H. J. Wang, J. H. Wang, Y. Xiao, and M. Q. Li, “Influence of surface cracks on laser-induced damage resistance of brittle KH₂PO₄ crystal,” Opt. Express 22(23), 28740–28755 (2014). [CrossRef]

27. 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]

28. J. Fournier, J. Neauport, P. Grua, E. Fargin, V. Jubera, D. Talaga, and S. Jouannigot, “Evidence of a green luminescence band related to surface flaws in high purity silica glass,” Opt. Express 18(21), 21557–21566 (2010). [CrossRef]

29. X. He, G. Wang, H. Zhao, and P. Ma, “Subsurface defect characterization and laser-induced damage performance of fused silica optics polished with colloidal silica and ceria,” Chin. Phys. B 25(4), 048104 (2016). [CrossRef]

30. T. Shao, Z. H. Shi, L. X. Sun, X. Ye, J. Huang, B. Li, L. M. Yang, and W. G. Zheng, “Role of each step in the combined treatment of reactive ion etching and dynamic chemical etching for improving the laser-induced damage resistance of fused silica,” Opt. Express 29(8), 12365–12380 (2021). [CrossRef]

31. E. Broitman, “Indentation Hardness Measurements at Macro-, Micro-, and Nanoscale: A Critical Overview,” Tribol. Lett. 65(1), 23 (2017). [CrossRef]

32. M. J. Chen, W. Y. Ding, J. Cheng, H. Yang, and Q. Liu, “Recent Advances in Laser-Induced Surface Damage of KH2PO4 Crystal,” Appl. Sci. 10(19), 6642 (2020). [CrossRef]

33. Y. L. Liu, L. Qi, M. K. Zheng, and R. Z. Zhang, “Effect of scratch defects on photoinduced damage characteristics of fused quartz,” Acta Phys. Sin. 70(13), 134203 (2021). [CrossRef]

34. V. A. Schanz, M. Roth, and V. Bagnoud, “Picosecond contrast degradation by surface imperfections in chirped-pulse-amplification stretchers,” J. Opt. Soc. Am. A 36(10), 1735–1742 (2019). [CrossRef]

35. R. D. Holbrook, P. C. DeRose, S. D. Leigh, A. L. Rukhin, and N. A. Heckert, “Excitation-emission matrix fluorescence spectroscopy for natural organic matter characterization: A quantitative evaluation of calibration and spectral correction procedures,” Appl. Spectrosc. 60(7), 791–799 (2006). [CrossRef]

36. S. M. Phillips and G. D. Smith, “Further Evidence for Charge Transfer Complexes in Brown Carbon Aerosols from Excitation-Emission Matrix Fluorescence Spectroscopy,” J. Phys. Chem. A 119(19), 4545–4551 (2015). [CrossRef]

37. C. Xu and W. W. Webb, “Measurement of two-photon excitation cross sections of molecular fluorophores with data from 690 to 1050 nm,” J. Opt. Soc. Am. B 13(3), 481–491 (1996). [CrossRef]

38. J. A. Gardecki and M. Maroncelli, “Set of secondary emission standards for calibration of the spectral responsivity in emission spectroscopy,” Appl. Spectrosc. 52(9), 1179–1189 (1998). [CrossRef]

39. R. Pedros, I. Moya, Y. Goulas, and S. Jacquemoud, “Chlorophyll fluorescence emission spectrum inside a leaf,” Photoch. Photobio. Sci. 7(4), 498–502 (2008). [CrossRef]

40. N. Chandrasekhar and K. Gartia, “Gaussian approximation of thermoluminescence (TL) peaks: A common misconception for analysis of TL of persistent luminescent materials,” J. Alloy. Compd. 745, 773–778 (2018). [CrossRef]

41. Q. Liu, J. Cheng, Z. R. Liao, X. C. Luo, Y. Yue, M. Q. Li, H. Yang, C. Tan, G. Z. Wang, W. Y. Ding, Z. Y. Yin, L. J. Zhao, and M. J. Chen, “Research on the light intensity modulation and characterizing methods of surface texture on KDP optics generated in fly-cutting and micro ball-end milling processes,” CIRP J. Manuf. Sci. Tec. 41, 30–43 (2023). [CrossRef]

42. B. J. Maciag and J. M. Brenan, “Speciation of arsenic and antimony in basaltic magmas,” Geochim. Cosmochim. Acta 276, 198–218 (2020). [CrossRef]

43. S. Bhardwaj, T. T. Fernandez, S. Gross, M. J. Withford, and M. J. Steel, “Quantitative morphology of femtosecond laser-written point-by-point optical fiber Bragg gratings,” Opt. Lett. 47(3), 453–456 (2022). [CrossRef]

44. C. H. Li, X. L. Kang, D. C. Luan, Q. H. Zhu, and W. G. Zheng, “Laser-irradiation-driven formation of oxygen-related defects and performance degradation in fused silica with nanosecond pulsed laser at 355 nm,” Opt. Laser Technol. 111, 727–733 (2019). [CrossRef]

45. E. A. Mikhantiev, I. G. Neizvestny, S. V. Usenkov, and N. L. Shwartz, “Monte Carlo simulation of the effect of silicon monoxide on silicon-nanocluster formation,” Semiconductors 48(7), 891–898 (2014). [CrossRef]

46. Y. D. Glinka, S. H. Lin, and Y. T. Chen, “Time-resolved photoluminescence study of silica nanoparticles as compared to bulk type-III fused silica,” Phys. Rev. B 66(3), 035404 (2002). [CrossRef]

47. K. Kajihara, L. Skuja, M. Hirano, and H. Hosono, “Role of mobile interstitial oxygen atoms in defect processes in oxides: Interconversion between oxygen-associated defects in SiO2 glass,” Phys. Rev. Lett. 92(1), 015504 (2004). [CrossRef]

48. B. Chimier, O. Uteza, N. Sanner, M. Sentis, T. Itina, P. Lassonde, F. Legare, F. Vidal, and J. C. Kieffer, “Damage and ablation thresholds of fused-silica in femtosecond regime,” Phys. Rev. B 84(9), 094104 (2011). [CrossRef]

49. J. H. Bin, M. Yeung, Z. Gong, H. Y. Wang, C. Kreuzer, M. L. Zhou, M. J. V. Streeter, P. S. Foster, S. Cousens, B. Dromey, J. Meyer-ter-Vehn, M. Zepf, and J. Schreiber, “Enhanced Laser-Driven Ion Acceleration by Superponderomotive Electrons Generated from Near-Critical-Density Plasma,” Phys. Rev. Lett. 120(7), 074801 (2018). [CrossRef]

50. B. Rethfeld, “Free-electron generation in laser-irradiated dielectrics,” Contrib. Plasma Phys. 47(4-5), 360–367 (2007). [CrossRef]

51. I. Boukhris, I. Kebaili, M. S. Al-Buriahi, A. Alalawi, A. S. Abouhaswa, and B. Tonguc, “Photon and electron attenuation parameters of phosphate and borate bioactive glasses by using Geant4 simulations,” Ceram. Int. 46(15), 24435–24442 (2020). [CrossRef]

52. X. F. Jing, J. D. Shao, J. C. Zhang, Y. X. Jin, H. B. He, and Z. X. Fan, “Calculation of femtosecond pulse laser induced damage threshold for broadband antireflective microstructure arrays,” Opt. Express 17(26), 24137–24152 (2009). [CrossRef]

53. M. Jupe, L. Jensen, A. Melninkaitis, V. Sirutkaitis, and D. Ristau, “Calculations and experimental demonstration of multi-photon absorption governing fs laser-induced damage in titania,” Opt. Express 17(15), 12269–12278 (2009). [CrossRef]

54. S. G. Demos, P. DeMange, R. A. Negres, and M. D. Feit, “Investigation of the electronic and physical properties of defect structures responsible for laser-induced damage in DKDP crystals,” Opt. Express 18(13), 13788–13804 (2010). [CrossRef]

55. B. Fritzke, J. Gotze, and J. M. Lange, “Cathodoluminescence of moldavites,” Meteorit. Planet. Sci. 52(7), 1428–1436 (2017). [CrossRef]

Evolution of the point defects involved under the action of mechanical forces on mechanically machined fused silica surfaces

Abstract

1. Introduction

2. Experimental section

2.1 Sample preparation

2.2 Defect preparation

2.3 PL-detection experiment

2.4 Laser-induced damage experiment

2.5 Scanning electron microscopy and 3D microscope measurement

3. Results and discussions

3.1 Determination of the point-defect types on mechanically machined optical surfaces

3.2 Relationship between the densities of oxygen-deficient and peroxide point defects

3.3 Relationship between PL, the densities of the unbonded electrons and the point defects

3.4 Experiment validation

4. Conclusion

Funding

Disclosures

Data availability

References

Data availability

Cited By

Figures (12)

Equations (16)

Optics Express