OSA's Digital Library

Optics Express

Optics Express

  • Editor: C. Martijn de Sterke
  • Vol. 20, Iss. 25 — Dec. 3, 2012
  • pp: 27708–27724
« Show journal navigation

Characterization of ejected fused silica particles following surface breakdown with nanosecond pulses

Rajesh N. Raman, Selim Elhadj, Raluca A. Negres, Manyalibo J. Matthews, Michael D. Feit, and Stavros G. Demos  »View Author Affiliations


Optics Express, Vol. 20, Issue 25, pp. 27708-27724 (2012)
http://dx.doi.org/10.1364/OE.20.027708


View Full Text Article

Acrobat PDF (5740 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

The light emission produced near the surface of fused silica following laser-induced breakdown on the exit surface was spatially and spectrally resolved. This signal is in part generated by ejected particles while traveling outside the hot ionized region. The thermal emission produced by the particles can be separated from the plasma emission near the surface and its spectral characteristics provide information on the temperature of the particles after ejection from the surface. Assuming the emission is thermal in origin, data suggest an initial average temperature on the order of at least 0.5 eV.

© 2012 OSA

1. Introduction

The nature of the material modifications caused by high average power laser irradiation is of great interest in various applications (such as laser micro-machining, laser-assisted deposition and laser-induced damage in high power laser systems) and represents a still not well-understood materials science topic. A key component in the interaction of laser pulses with materials during laser-induced breakdown is the ejection of particles (material clusters). These particles can lead to a number of adverse effects in various applications, including degradation of coating performance in laser-assisted deposition. They are also of significant concern in the case of laser damage in optical components employed in large aperture, high average power laser systems for laser–driven inertial confinement fusion (ICF), mainly as a secondary source of contamination of adjacent optics leading to new sites of damage initiation under subsequent exposure to laser pulses. Such systems are designed to be able to operate while their optical components have sustained some laser-induced damage [1

1. A. K. Burnham, L. Hackel, P. Wegner, T. Parham, L. Hrubesh, B. Penetrante, P. Whitman, S. Demos, J. Menapace, M. Runkel, M. Fluss, M. Feit, M. Key, and T. Biesiada, “Improving 351-nm damage performance of large-aperture fused silica and DKDP optics,” Proc. SPIE 4679, 173–185 (2002). [CrossRef]

].

The optical emission generated during laser-induced breakdown has been employed in the past to determine specific characteristics of the generated plume [2

2. S. S. Harilal, C. V. Bindhu, R. C. Issac, V. P. N. Nampoori, and C. P. G. Vallabhan, “Electron density and temperature measurements in a laser produced carbon plasma,” J. Appl. Phys. 82(5), 2140–2146 (1997). [CrossRef]

7

7. C. Aragón and J. A. Aguilera, “Characterization of laser induced plasmas by optical emission spectroscopy: a review of experiments and methods,” Spectrochim. Acta B 63(9), 893–916 (2008). [CrossRef]

]. This emission originates primarily from electron-ion interactions (yielding both continuum and discrete spectra) and electron-laser interactions (in the case of nanosecond and longer pulses) [8

8. J. P. Singh and S. N. Thakur, eds., in Laser-Induced Breakdown Spectroscopy (Elsevier, Oxford, 2007).

]. For example, the electron temperature, electron density, ion temperature, pressure, and plume expansion speed have been characterized using the spectral and temporal characteristics of the emission.

The spectrally resolved optical emission of nanoparticles generated under ultrafast pulsed laser ablation of Si surfaces has been examined to estimate the particles’ thermal characteristics. By assuming their emission is thermal and near blackbody in nature, it was suggested that the particle temperature is on the order of 2300 K [15

15. S. Amoruso, R. Bruzzese, N. Spinelli, R. Velotta, M. Vitiello, X. Wang, G. Ausanio, V. Iannotti, and L. Lanotte, “Generation of silicon nanoparticles via femtosecond laser ablation in vacuum,” Appl. Phys. Lett. 84(22), 4502–4505 (2004). [CrossRef]

]. In the field of laser damage, particles are ejected from the material volume exposed to laser energy deposition and they traverse the plasma generated as part of the process. In recent work, we investigated the temporal evolution and kinetic properties of material (micron-size) particles ejected from the surface of fused silica under nanosecond laser irradiation using a time-resolved microscope system [16

16. R. N. Raman, R. A. Negres, and S. G. Demos, “Kinetics of ejected particles during breakdown in fused silica by nanosecond laser pulses,” Appl. Phys. Lett. 98(5), 051901 (2011). [CrossRef]

]. These experiments demonstrated that there is a spectrum of particle sizes from 1 µm in diameter (detection limit of instrumentation) up to about 50 µm or larger. These particles travel with various speeds, with the early ejected particles (which are smaller in size) having speeds on the order of 2-3 km/s. The shape of some of the ejected particles having lower speeds suggests that they are the result of mechanically damaged material. However, the faster particles give the impression of a material explosively ejected from a low viscosity molten volume. The particle ejection process has duration on the order of 10 µs, and near the end of the process the ejecta speed is reduced to about 10 m/s or less. It is assumed that early ejected particles are in a liquid phase since they are the product of the host material response to extreme temperature and pressure conditions.

In this work, the broadband emission partially arising from material ejected during localized laser energy deposition on the exit surface of fused silica optical windows was captured with the aid of temporally-integrated imaging and spatially-resolved emission spectroscopy. The experimental system is able to spatially resolve the emission components arising from the ionized region closer to the surface from the emission of the hot ejected particles while they are traveling away from the surface. The laser energy is deposited on the exit surface of the optic (as opposed to the front surface typically associated with laser ablation and machining applications) which prevents secondary energy deposition from the laser pulse into the developing plume. Instead, the modified material front expands into the bulk without any obstruction of the excitation laser pulse, and likewise the ensuing material response including the ejection of particles is not influenced by secondary processes. The observations are interpreted in conjunction with information on the position, speed, and trajectory of the ejected particles during the breakdown event as measured by time-resolved shadowgraphic microscopy and time-integrated light scattering imaging. The results suggest that the spectral characteristics of the emission vary with increasing distance from the surface. The origin of this emission is assigned to three main components which dominate the signal at different positions away from the surface: i) ionized material, ii) faster, hotter, and relatively smaller particles and iii) slower, cooler, and relatively larger particles.

2. Experimental methods

Breakdown was induced on the exit-surface of a 5-cm round, 1-cm thick UV grade Type III fused silica optical window (CVI Melles-Griot, laser-grade polish) by focusing a laser pulse (henceforth referred to as pump pulse) about 2 cm behind the optic’s exit-surface (see schematic in Fig. 1(a)
Fig. 1 Schematic of experimental arrangements. (a) Time-resolved and time-integrated imaging setup. L = focusing lens, M = mirror, PBS = polarizing beam-splitter. (b) Spatially-resolved emission spectroscopy setup.
). The output from different laser systems was used to initiate breakdown. The first laser was operating at 1064 nm with pulse duration of 10-ns full-width at half maximum (FWHM) and an average fluence of the converged beam at the exit surface on the order of 200 J/cm2 by means of a 5 cm focal length lens. The second laser was operating at 355 nm with pulse duration of 7.5-ns FWHM and an average fluence of the converged beam at the exit surface on the order of 100 J/cm2 via a 15 cm focal length lens. The qualitative results regarding the speed and size of the ejected particles following laser breakdown were found to be essentially identical for these two pump lasers. However, the amount of material ejected is very sensitive to the laser fluence as its increase leads to a larger volume of ejected material. The damage craters generated were on the order of 150 µm in diameter having a depth on the order of about 30 µm and 40 µm for the 1064 nm and 355 nm laser pulses, respectively.

The kinetic properties of the particles were estimated by shadowgraphic imaging using a time-resolved microscope system described in detail elsewhere [17

17. R. N. Raman, R. A. Negres, and S. G. Demos, “Time-resolved microscope system to image material response following localized laser energy deposition: exit surface damage in fused silica as a case example,” Opt. Eng. 50(1), 013602 (2011). [CrossRef]

]. In brief, the side of the sample was illuminated with two spatially overlapped, orthogonally polarized 532 nm, 4.5 ns FWHM probe laser pulses, as illustrated in Fig. 1(a) (green beam paths). Each probe pulse was delayed with respect to the peak of the pump pulse via adjustable external triggers, allowing side-view illumination of the sample at two time points during a single laser energy deposition event. A composite 5X zoom and 5X objective lens system was used to collect the dual-probe signal traversing the ejected material volume which subsequently passed through a 532-nm narrowband filter and was sorted using a polarizing beam-splitter. Images from each probe beam were recorded by separate charge-coupled device (CCD) cameras. Static optical resolution was better than 2 µm with a depth of focus of ~40 µm. Particle average velocity was then estimated from the distance traveled during the time interval between the two probe pulses.

The time-resolved imaging method described above was complemented by two additional imaging approaches using the same basic system depicted in Fig. 1(a). The first approach is based on time-integrated light scattered from the particles under laser illumination from a continuous-wave, 670 nm diode laser. These images capture the light scattered by the particles during a single breakdown event, and therefore the intensity is dominated by signal originating from the slower moving particles. Consequently, this method allows the acquisition of information on the trajectories of the slowest particles which cannot be studied by the time resolved imaging method (more suitable for capturing snapshots of fast events). The light scattering images were captured by the CCD camera denoted as CCD #1 in Fig. 1(a) after removing the polarization element (denoted as PBS) located after the microscope’s imaging optics while the probe lasers were turned off and the 532-nm narrowband filter was replaced by a 670-nm narrowband filter.

The second imaging method is based on the light emission produced near the location of laser breakdown under the same damage conditions described above. The same time-integrated imaging configuration was used but the 670-nm laser was turned off and the corresponding narrowband filter removed. The emission image was therefore based on spontaneously generated photons in the 400-800 nm spectral range (limited by the transmission window of the optical elements involved and wavelength sensitivity of the CCD camera chip). This method was used to resolve the spatial distribution of the emission generated during and following energy deposition.

In addition to the imaging approaches described above, a spectroscopy system was used to spectrally and spatially resolve the light emission signal. For that purpose, the area near the surface of the sample was imaged using a 50 mm camera lens with magnification of 0.89 onto a linear array fiber bundle (16 fibers, each having a 250 µm diameter) as shown in Fig. 1(b). The linear fiber array was oriented perpendicular to the sample’s surface and along the direction of material ejection. Consequently, each fiber captures the signal produced over a cylindrical volume of 280 µm in diameter (250 µm/0.89) and on the order of 1000 µm in length.

Overall, the fiber array collected the emission from a region extending up to about 4.2 mm from the sample’s exit surface. The fiber output was coupled to the entrance slit of a spectrometer equipped with a diffraction grating blazed at 450 nm and imaged onto a liquid-nitrogen cooled CCD after being spectrally resolved. The sample was translated laterally during a 4 minute signal acquisition, such that a pristine region was exposed to each pulse from the pump laser operating at 10 Hz. This spectral acquisition integration time was used to optimize the signal from the fibers probing the most distant locations with respect to the surface without saturating the CCD pixels exposed to signal from the fibers located near the surface. The spectrum of the emission collected by each fiber was obtained by averaging the signal over 16 pixels in height centered on each fiber to include signal coming from the core. The background was measured by collecting signal with the sample translated away from the focal point of the lens so that the laser fluence on the surface is below the breakdown threshold. This background contained dark noise signal and a small component of the air breakdown signal produced near the focal point of the lens focusing the pump laser beam. Although the air breakdown region was outside the field imaged onto the fiber array, some stray light was reaching the fibers after reflections at the optical table or other optical elements. Subtractions of the background signal as well as corrections for the instrument response were subsequently performed on the spectra from individual fibers.

3. Results

3.1. Experimental results

The side-view shadowgraphic images shown in Figs. 2
Fig. 2 Images of damage events on the exit surface (located on the right hand side) of fused silica. (a) Shadowgraphy images taken at 1 µs (a1) and 20 µs (a2) delays, respectively. Time-integrated (b) broadband emission and (c) 670-nm light scattering images acquired during a single event (log intensity scale). All images have the same spatial scale.
(a1) and 2(a2) reveal the location of ejected particles following a breakdown event at 1 µs and 20 µs delay, respectively. A jet of particles has been formed in which the smaller, faster particles (< 5 µm, ~1-2 km/s) on average have traveled further away from the surface, while larger, slower, irregularly-shaped clusters (≥15 µm, <200 m/s) can be seen closer to the surface and at later delays. The particle jet is concentrated along a direction perpendicular to the surface. The largest particles have a diameter on the order of 30-50 µm but particles as large as about 100 µm in diameter can be occasionally observed (although those are typically flakes with very high aspect ratio). These larger particles are also slower, having speed on the order of 100 m/sec or less. The speed of the particles is indicated in Fig. 2(a1) separating the particles in three groups having high speed (larger than 1 km/s), intermediate speed (0.4 to 1 km/s) and lower speed (less than 0.4 km/s).

The ejection of particles forming a relatively confined jet is confirmed by the light scattering images such as the typical example shown in Fig. 2(c). This image is displayed using a log intensity scale in order to enhance visibility of the particles ejected with higher speeds and thus produce a relatively smaller time-integrated signal. These fast particles are best visualized with the shadowgraphy microscope system as discussed above and give rise to most of the low intensity features observed in Fig. 2(c). In contrast, the slow particles ejected at the end of the process which have very low speeds produce stronger integrated light scattering signal, and are observed in Fig. 2(c) as the brighter features with well-defined trajectories. The trajectory of some of these particles is observed to curve back towards the surface. The change in direction of the slow moving particles towards the surface may in part be attributed to two mechanisms. First, following the initial shockwave expansion and ensuing rarefaction wave produced by the microexplosion, a compression wave associated with the air drawn back into the surface is formed. The drag forces generated on smaller and slower particles can cause change in their trajectory [18

18. G. Koren and U. P. Oppenheim, “Laser ablation of polymers in pressurized gas ambients,” Appl. Phys. B 42(1), 41–43 (1987). [CrossRef]

20

20. F. Wagner and P. Hoffmann, “Structure formation in excimer laser ablation of stretched poly(ethylene therepthalate) (PET): the influence of scanning ablation,” Appl. Phys., A Mater. Sci. Process. 69(7), S841–S844 (1999). [CrossRef]

]. The second possible mechanism may be associated with interaction of the debris with electrostatic forces which are known to be generated on the surface of dielectrics following laser radiation [21

21. J. T. Dickinson, S. C. Langford, J. J. Shin, and D. L. Doering, “Positive ion emission from excimer laser excited MgO surfaces,” Phys. Rev. Lett. 73(19), 2630–2633 (1994). [CrossRef] [PubMed]

,22

22. S. R. George, J. A. Leraas, S. C. Langford, and J. T. Dickinson, “Interaction of vacuum ultraviolet excimer laser radiation with fused silica. I. Positive ion emission,” J. Appl. Phys. 107(3), 033107 (2010). [CrossRef]

]. As only a small portion of the area of the sample’s surface exposed to the laser beam gives rise to ablation, there is a relatively larger area surrounding the site of material ejection that is exposed to a high fluence irradiation similar to the conditions discussed in Refs. 21

21. J. T. Dickinson, S. C. Langford, J. J. Shin, and D. L. Doering, “Positive ion emission from excimer laser excited MgO surfaces,” Phys. Rev. Lett. 73(19), 2630–2633 (1994). [CrossRef] [PubMed]

and 22

22. S. R. George, J. A. Leraas, S. C. Langford, and J. T. Dickinson, “Interaction of vacuum ultraviolet excimer laser radiation with fused silica. I. Positive ion emission,” J. Appl. Phys. 107(3), 033107 (2010). [CrossRef]

that can lead to the generation of electrostatic forces. This issue will be discussed in more detail at the end of this section.

Figure 2(b) shows a typical image of the spatial distribution of the emission signal (log intensity scale) produced in the region near the surface breakdown location and captured over the course of a single event. This emission “fireball” exhibits a wider angle of divergence compared to the angle of particle ejection (the jet of particles) observed in the snapshots captured in Fig. 2(a1) and 2(a2). The intensity is decreased to below the detection limit of our 12-bit CCD detector at a distance of about 1.25 mm from the surface. This is much smaller than the typical plasma plume produced by front surface laser breakdown (typical geometry used in laser ablation of materials) where additional laser energy is absorbed in the expanding plume.

Figure 3(a)
Fig. 3 (a) Normalized spectra from various fibers integrated over 2400 ablation events. (b) Peak intensity and corresponding position (in wavelength) of measured spectra for all fibers.
shows the emission spectra in the 400-750 nm spectral range (normalized to their peak intensity) obtained by the fiber array. Each fiber represents the signal collected over a cylindrical volume with radius of 280 µm as discussed earlier. The system was arranged so that fiber 0 (spectrum not shown) collects the signal from the surface while fiber 1 collects the signal at a distance of about 280 µm from the surface, with subsequent fibers covering a distance up to about 4.2 mm from the surface of the sample. The values of the peak emission intensity and the corresponding wavelength from spectra collected from each fiber are displayed in Fig. 3(b) (right vertical axis and left vertical axis, respectively). The spectra from fiber 5 and beyond are broad with no well-defined peaks. The spectra of fibers 1-4 exhibit twopeaks at about 510 nm and 635 nm. The 510 nm peak is also observed during air breakdown. Assuming that these peaks arise from the formation of plasma and accompanying atomic emission lines by the basic elements (nitrogen, oxygen and silicon), the 510 nm peak can be assigned to emission from nitrogen [23

23. V. Narayanan, V. Singh, P. K. Pandey, N. Shukla, and R. K. Thareja, “Increasing lifetime of the plasma channel formed in air using picoseconds and nanosecond laser pulses,” J. Appl. Phys. 101(7), 073301 (2007). [CrossRef]

] while the 635 nm peak can be assigned to emission from silicon [24

24. A. Huber, I. Beigman, D. Borodin, P. Mertens, V. Philipps, A. Pospieszczyk, U. Samm, B. Schweer, G. Sergienko, and L. Vainshtein, “Spectroscopic observation of Si I- and Si II- emission lines in the boundary of TEXTOR and comparison with kinetic calculations,” Plasma Phys. Contr. Fusion 45(2), 89–103 (2003). [CrossRef]

]. The spectral profile shifts toward shorter wavelengths with increasing distances from the exit surface up to fiber 5, which coincides with the distance from the surface that is at the outer boundary of the emission “plasma fireball” shown in Fig. 2(b). The spectra from fibers 6 through 8 appear to develop a second spectral band centered at about 570 nm which becomes the dominant feature from fiber 9. Moreover, this latter feature appears to red-shift to longer wavelengths with increasing distance from the surface (see spectra from fiber 7 to fiber 14 in Fig. 3(a)).

There are a wide range of particle sizes and speeds produced during a laser breakdown event under the experimental conditions used in this work [16

16. R. N. Raman, R. A. Negres, and S. G. Demos, “Kinetics of ejected particles during breakdown in fused silica by nanosecond laser pulses,” Appl. Phys. Lett. 98(5), 051901 (2011). [CrossRef]

]. This effect is also captured in the time-resolved shadowgraphy microscopic images during material ejection shown in Fig. 2. The fastest visible ejected particles have a speed on the order of 2 km/s and a diameter on the order of 5 µm or less. Thereafter, the particle speed is reduced while particles with larger sizes are present. To help interpret the experimental results shown in Figs. 3(a)-3(b) for distances outside the plasma region, we acquired time resolved images in the range between 3 mm and 4.5 mm distance from the surface corresponding to the region interrogated by fibers 10 to 15 in the experimental system shown in Fig. 1(b). These images were acquired at different delay times to capture particles having different average speeds. Typical results are provided in the images shown in Fig. 4
Fig. 4 Time-resolved images capturing the spatial and size distributions of ejected particles at a distance between 3 - 4.5 mm from the surface at (a) 4.5 µs, (b) 10 µs, (c) 20 µs, and (d) 50 µs delays. These delays capture ejected particles having different average speeds as denoted in the lower part of each image. The inset to the left in (a) depicts a magnified image of a small particle, while the inset to the right depicts a magnified image of a segment of the shock wave. More details are provided in the text.
obtained at a) 4.5 µs, b) 10 µs, c) 20 µs and, d) 50 µs delays from the pump pulse, respectively. Specifically, Fig. 4(a) captures particles having average speeds (determined from the distance traveled divided by the delay time) of over 500 m/s. The particles in this speed range are rather small, having diameter not exceeding about 10 µm. The particles with higher speeds are smaller (less than 5 µm in diameter) and 6 of these are identified in the image by red arrows. One of these small particles is shown with higher magnification as inset on the left. The location of the shock (pressure) wave is also visible at a distance of about 3.8 mm from the surface and is denoted with a single blue arrow. A small section of the image portion containing the shockwave (located above the point shown by the arrow) is shown as inset with higher magnification and enhanced contrast to improve itsvisualization. Particles that are out of focus are also visible presenting characteristic ring patterns that help not to be misinterpreted as larger particles. Figure 4(b) shows particles having speeds between about 500 m/s and 250 m/s. It can be appreciated that in this range of speeds, in addition to the small particles there are larger particles having diameter up to about 30 µm. The particles are quasi-spherical and are assumed to be the product of the explosive process leading to the material ejection following the rapid heating of the material by the laser (pump) pulse. A similar behavior is depicted in Fig. 4(c) capturing particles having speeds between about 250 m/s and 125 m/s having diameter up to about 50 µm. Figure 4(d) captures the image of particles having speeds between about 100 m/s and 50 m/s and indicate the presence of very large material “flakes” (2 of which are indicated by green arrows), which have also been discussed in Ref. 16

16. R. N. Raman, R. A. Negres, and S. G. Demos, “Kinetics of ejected particles during breakdown in fused silica by nanosecond laser pulses,” Appl. Phys. Lett. 98(5), 051901 (2011). [CrossRef]

. The flakes are most likely the product of mechanical damage arising from the stresses and pressure waves produced by the initial explosive material ejection process. We can therefore postulate that their initial temperature was not very high or at nearly room temperature in contrast with the particles having higher speeds that were part of the overheated material following laser energy deposition.

As discussed earlier, Fig. 2(c) demonstrates the presence of particles that change their direction of propagation after ejection. It was hypothesized that this process may be attributed to either the interaction of slow moving particles with the air flow after the termination of the expansion of the shockwave or the buildup of electrostatic forces. Additional experimental results aiming at elucidating this issue are presented in Fig. 5
Fig. 5 (a) Time-integrated light scattering images under 670-nm CW laser illumination acquired during a single event (linear intensity scale). (b) Trajectories of particle paths exhibiting change in propagation obtained from 7 different images, with each color representing a different image.
. Specifically, another example of a time-integrated light scattering image under 670-nm CW laser illumination is shown in Fig. 5(a) displayed in linear intensity scale. This selected image is dominated by the trails of the light scattering signal of two types of particles as they travel through the imaged area and while the camera is turned on. The first type is associated with high aspect ratio large particles (flakes) that travel in a straight line while rotating giving rise to the observed helical tracks.

Such flakes are also observed in Fig. 4(d) and have speeds on the order of 50-100 m/s. The second type is associated with particles that change trajectory after ejection. To obtain a better understanding of the processes involved in changing the direction of propagation of theseparticles, Fig. 5(b) summarizes the digitally acquired trajectories of such particles as recorded in multiple images. The color coding is used to track particles observed during the same events, i.e., trajectories of the same color represent particles observed in the same image. These particles have a diameter on the order of 10 to 20 µm (estimated from the width of the trajectory lines). Based on the brightness of these particle trajectories compared to the rest of the scattering features arising from the faster particles and our understanding of the speed distribution of the ejected particles, we deduce that the speed of these particles is on the order of 10 m/s or less. These trajectories appear to have approximately the same intensity during the propagation of the particle. Since the brightness of the detected signal by the camera for each trajectory is inversely proportional to the speed of the particle (the time spent by the particle in each image segment), it can be concluded that the average speed of the particles does not significantly change during the propagation. It must be noted that the surface of the material is positioned vertically and therefore, gravity is not influencing the observed changes in trajectory.

The particle trajectories shown in Fig. 5(b) indicate an initial nearly linear propagation followed by an abrupt change in direction. These particles travel a distance on the order of 250 to 500 µm during which their direction of motion is changing after which the particles appear to return to a mostly linear propagation. Therefore, the presence of a force(s) applied on the particles is not continuous and initiates after the particles have traveled a distance between about 300 to 800 µm from the surface. As such behavior is very difficult to be reproduced assuming only electrostatic forces, we postulate that it may arise from aerodynamic forces developed near the surface at some point following the laser-induced breakdown event. Following laser energy deposition, the temperature and pressure of the heated gas is lowered in part as a result of the expansion. As the pressure of the expanding gas drops to a value on the order of atmospheric, the difference in density between the surrounding cold and hot region causes their mixing with an inflow of cold gas into the hot region. This flow of gas will reach the region near the location of laser induced breakdown at a certain delay from the time of initial energy deposition. We hypothesize that the change in the direction of the slower particles may be the result of their interaction with this air flowing back toward the surface. The flakes seen in Fig. 5(a) do not seem to encounter this air front within the distance of at least 1.5 mm from the surface. Given that the speed of the slower flakes is on the order of 50 m/s, the delay of the arrival of this air front should be longer than about 30 µs. However, for the slower and smaller particles with a speed on the order of 10 m/s or less, the air front will intercept the particles while they have traveled some distance from the surface causing an abrupt change in the trajectory. Assuming from trajectories shown in Fig. 5(b) that the particles first traveled about 300 µm from the surface having a nearly linear trajectory and assuming a speed of initial propagation of about 5 m/s, the delay for the arrival of the air front is estimated to be on the order 60 µs. It must be noted that these particles are also ejected with some delay with respect to the pump pulse which is on the order of 10 µs or longer.16 The faster of these slow particles will intercept the air front at a larger distance from the surface and their trajectory will be affected less than that of the slower particles, in agreement with the results shown in Fig. 5(b). We can estimate the magnitude of the air inflow velocity by calculating the finite time deceleration g necessary to bend the trajectories in the manner seen experimentally. This implies a force on the particle due to the differential pressure ΔP given by (4/3)ρpga where a is the particle radius and ρp its density. The net velocity of the air is then νair43(ρp/ρair)ga. Assuming a 15 µm diameter and typical geometric values for the turn-around points and initial velocities as shown in Fig. 5(b), estimated air speeds of order 10-100 m/s were found. These values correspond to differential pressures between cold and hot gas of up to 0.2 atm.

It must be noted that the excitation geometry used in this experiment involves ablation via a laser pulse that impinges on the surface from within the material (exit surface excitation). This geometry is distinctly different from the classical front illumination geometry typically used in laser ablation applications. As a result, the laser energy is deposited via the formation of an absorption front near the surface that propagates away from the surface and towards the bulk of the material. In addition, the amount of gaseous plume produced is much lower while a potentially larger number of ejected particles is created. This excitation geometry is particularly suitable for understanding the material behavior following energy deposition without the influence of the laser-plume interaction that is postulated to play a key role in theplume expansion dynamics in the case of front illumination geometry while limiting the amount of laser energy deposited in the material. In addition, the excitation geometry used in this work is directly relevant to laser- induced damage on the exit surface of optical components under high power nanosecond pulses which is a particularly important issue in large aperture laser systems.

Based on the above experimental observations, we can categorize the particles on the basis of their speeds and sizes. The first (fastest) group is represented by the ejected particles on the left hand side of Fig. 2(a1) and Fig. 4(a). The particles in this group are faster and generally small having diameters of less than about 10 µm. The particles in the middle section of the image of Fig. 2(a1) and Figs. 4(b)-4(c) represent the second group which contains slower particles but some are also larger, up to about 50 µm (or more) in diameter. Both of these groups are comprised of particles having nearly spherical or elliptical shape and it is presumed that they originate from the originally superheated material. The third group of particles is on the right hand side of the image in Fig. 2(a1) and all particles in Fig. 2(a2) as well as in the image of Fig. 4(d). This group of particles has speeds of less than 100 m/s and contains at least in part mechanically damaged material (such as flakes) that were not significantly heated by the laser energy deposition process. We will use this categorization of particle groups in the following modeling section aiming to, at least qualitatively, explain the experimental observations. We hypothesize that an understanding of the contribution to the thermal emission as a function of size and speed of the responsible particles can help interpret the experimental results shown in Fig. 3.

3.2. Model and simulation results

As noted earlier, laser-induced breakdown on the exit surface of silica is a complex process involving an array of interacting processes developing simultaneously. Based on the background information discussed in the introduction, the speed of the particles is lower than the initial speed of plasma expansion, and therefore, the faster particles will traverse from the surface in the presence of the hot plasma (inside the emission “fireball” shown in Fig. 2(b)). During that period of time, evaporative cooling would be greatly reduced, thus the temperature of the particles will not change significantly. The plasma will affect the first and second particle groups as defined above. The slow (group 3) particles, however, will not be affected significantly by this effect, as the plasma temperature will decay before these particles traverse a significant distance from the surface.

To simplify this problem, we will consider only the behaviors observed outside that plasma region which is characterized by the absence of atomic line emission (as discussed above) suggesting that most of the observed emission arises from the thermal radiation of the ejected particles. Our modeling approach is two-fold. First, we use a rate equation model to describe the energy balance during the cooling of the particles without taking into consideration the heat diffusion within each particle. This approach can describe the small particles in which the heat diffusion length (within the relevant time scales which depend on the speed of the particles) exceeds the radius of the particle. This approach, however, would not accurately describe the larger and slower particles of group 2 where energy loss will be dominated by the surface temperature while the temperature at the center may be higher. For this reason, we also employ a commercially available simulation software package (COMSOL Multiphysics) to solve the heat flow equation applied to a single spherical particle in air with the surface heat fluxes (described below) as boundary conditions. The results are used to model the behavior of the larger particles and to compare predictions for smaller particles with numerical results. Assuming that the spectra outside the plasma region shown in Fig. 3 arise from thermal emission by the ejected material and that this emission is near blackbody in nature, the position of peak intensity of the spectrum provides a first degree approximation regarding the average temperature of the particles dominating the recorded signal. As the emitted radiation is proportional to the temperature to the fourth power, it is evident that the hottest particles dominate the detected signal. In this context, the peak emission of about 430 nm recorded in fiber 6 (about 1.7 mm from the surface) corresponds to a temperature (through Planck’s relation) of about 6700 K, and a peak emission of about 560 nm to 580 nm recorded in fiber 10 to 15 (about 2.8 mm to 4.2 mm from the surface) corresponds to a temperature of about 5200 K to 5000 K.

To aid in understanding the results, we first used a simple model including radiative cooling and evaporative mass and energy loss. This model was formulated in terms of spherical particles of radius a. We further assumed no internal temperature variation within the particle and that radiative losses are blackbody with constant emissivity. The particle has initial temperature T0 and initial radius a0. During cooling, the change in particle temperature and size are determined by the coupled mass and energy conservation equations:

13ρCpda(t)3T(t)dt=a(t)2[(σ(T(t)4Ts4)+VmEhaΔTa(t))]13ρda(t)3dt=a(t)2Vm
(1)

Here ρ is the silica density (2200 kg/m3), Cp is silica specific heat 1400 (J/kg∙K), a is the particle radius, T is particle temperature, Ts is the surrounding radiation temperature, σ is the Stefan-Boltzmann constant, ha is the forced convection heat transfer coefficient taken to be ≈100 W/m2K since no values are available for the temperature, composition, and configuration in our experiment. Since convection contributes <2% of the total particle heat dissipation, this approximation does not produce any significant errors in the calculated temperatures. E is initially taken as equal to the evaporative molar energy in air ΔHr-CpT involved with the thermal decomposition of SiO2 to form SiO [27

27. H. L. Schick, “Thermodynamic analysis of the high temperature vaporization properties of silica,” Chem. Rev. 60(4), 331–362 (1960). [CrossRef]

]. The temperature dependence of ΔHr = −7.52 × T(t) + 1.94 × 105 (cal/mol) is from Ref [27

27. H. L. Schick, “Thermodynamic analysis of the high temperature vaporization properties of silica,” Chem. Rev. 60(4), 331–362 (1960). [CrossRef]

]. The assumption of uniform temperature within the particle is reasonable for particles smaller than their thermal diffusion length. Using fused silica thermal properties, this length corresponds to about 15 µm diameter particles for a period of 10 µs. A more correct treatment of cooling of larger particles is described below.

Evaporation is the dominant energy loss mechanism for temperatures above the boiling point of silica (3000 K) where the vapor pressure is 1 atm [27

27. H. L. Schick, “Thermodynamic analysis of the high temperature vaporization properties of silica,” Chem. Rev. 60(4), 331–362 (1960). [CrossRef]

]. In this regime, the temperature drops rapidly although the size decreases modestly over the time scale considered. Below 2500 K, radiative cooling dominates, the particle size does not change significantly and the temperature continues to drop. This process is depicted in Fig. 6
Fig. 6 Modeling of the temperature during cooling for particles with initial radius of a = 2.5 µm and a = 10 µm and initial temperatures of 6000 K assuming propagation in vacuum or air.
for the two different particle sizes having radius of a = 2.5 µm and a = 10 µm (group 1 and group 2) and initial temperatures of 6000 K. For comparison, two sets of evaporative mass flux and enthalpy (Vm(T) and E, respectively) will be used in Eq. (1) corresponding to a) evaporation of silica in vacuum and b) evaporation in ambient air. Specifically, using evaporation parameters for the case where the ejected particles are traveling in vacuum (evaporation parameters Am = 8.36 × 103 μg/μm2∙s and Bm = 82.8 kcal/mol from Ref. [26

26. S. I. Anisimov and V. A. Khokhlov, Instabilities in Laser-Matter Interaction (CRC Press, 1999).

]) as the basis for the model in Eq. (1), the profiles shown in Fig. 6 represent the calculated particle temperature as a function of time during cooling. The results demonstrate a much faster cooling for the smaller (a = 2.5 µm) particles for which the evaporation cooling process is terminating in less than about 60 µs (temperature is reduced to less than about 2500 K) when the rapid reduction in particle size terminates. In contrast, this temperature is reached by the larger (a = 10 µm) particles at about 240 µs from the onset of the cooling process. These differences are due to the larger surface-to-volume ratio for the smaller particles since both the evaporative and radiation losses are scaled linearly by the particle surface area.

The particle cooling model above can be used to interpret the temperature measurements of the particles as they traverse the region interrogated by each fiber. As discussed earlier, theplasma provides a source of energy for the particles [9

9. P. S. Dalyander, I. B. Gornushkin, and D. W. Hahn, “Numerical simulation of laser-induced breakdown spectroscopy: modeling of aerosol analysis with finite diffusion and vaporization effects,” Spectrochim. Acta B 63(2), 293–304 (2008). [CrossRef]

11

11. V. Hohreiter and D. W. Hahn, “Plasma-particle interactions in a laser-induced plasma: implications for laser-induced breakdown spectroscopy,” Anal. Chem. 78(5), 1509–1514 (2006). [CrossRef] [PubMed]

], thus the cooling phase of the faster particles (groups 1 and 2) will start only after they exit this hot ionized volume. Figure 2(b) indicates that in our case the plasma volume extends at least up to 1.25 mm from the surface (corresponding to fiber 5). The time for each particle to traverse the interrogation volume of each fiber depends on its speed. Specifically, a particle moving with a speed of about 1 km/s covers this distance in about 0.3 µs while a particle having speed of 100 m/s requires 3 µs. Based on the results shown in Fig. 6, the particles with radius of 2.5 µm will cool to a temperature of less than about 5000 K in a time less than that to traverse the distance of 10 fibers thus exhibiting a change of temperature of about 100 K as it transverses each fiber. For the particles with radius of 10 µm, the temperature declines more slowly and remains above 5000 K for about 10 µs which for a particle with a speed of about 500 m/s corresponds to the time needed to traverse about 8 fibers. This also yields a change in temperature of about 125 K as it transverses each fiber. In both case examples presented above involving the smaller and faster particles, the spectrum of the thermal emission of the particles will continuously red-shift as it transverses the fibers and the total contribution to the recorded signal will rapidly decline, as it depends on the fourth power of the particle’s temperature.

The above estimates suggest that the smallest particles will contribute more to the thermal signal collected by the fibers just outside the plasma zone (corresponding to fibers 5 and beyond) and should contribute to a continuously red-shifting signal component as a function of the fiber distance from the edge of the plasma zone. To understand the origin of the spectral component centered at about 570 nm in fiber 10 and above we need to understand the cooling of the larger particles of group 2 as mentioned above, when the kinetics of the intra-particle heat transfer become significant. The COMSOL Multiphysics platform was used to include thermal transport inside the heated particle in the theoretical description of particle cooling. Particle cooling from an initial uniform temperature through radiation, evaporation and convection was simulated using arbitrary Lagrangian-Eulerian finite element analysis of the heat flow equation:

ρCp(T)Tt[k(T)T]=Q
(2)

In Eq. (2), ρ, Cp and k are the density, specific heat capacity under constant pressure and the thermal conductivity respectively. In terms of heat transport, ρ varies little over the temperature range of interest and is taken as constant ρ = 2.2 g/cm3 [29

29. R. Brückner, “Properties and structure of vitreous silica. I,” J. Non-Cryst. Solids 5(2), 123–175 (1970). [CrossRef]

]. However, Cp and k vary appreciably over temperature and were modeled using published data for Corning 7980 glass [30

30. S. T. Yang, M. J. Matthews, S. Elhadj, V. G. Draggoo, and S. E. Bisson, “Thermal transport in CO2 laser irradiated fused silica: in situ measurements and analysis,” J. Appl. Phys. 106(10), 103106 (2009). [CrossRef]

]. The heat source term, Q, is set as a boundary condition and is a sum of evaporative, convective (h = 100 W/m2/K) and radiative (emissivity = 1) losses as described above. A 2D axisymmetric domain was used with >11,000 moving mesh elements each of which with a maximum 10 nm length. Heat transport and the deformed geometry were fully coupled and solved using a staggered, operator split method [31

31. F. Armero and J. C. Simo, “A new unconditionally stable fractional step method for nonlinear coupled thermomechanical problems,” Int. J. Numer. Methods Eng. 35(4), 737–766 (1992). [CrossRef]

]. At each time step a prescribed surface velocity derived from Vm(T) was calculated allowing the geometry to deform and mesh elements to evolve using Laplace smoothing. In this way, the moving particle surface is treated self-consistently with heat flow as material evaporates away. Solutions to Eq. (1) and Eq. (2) agree (not shown) as expected for longer cooling time when heat losses become much slower, and for smaller particle sizes comparable to the thermal diffusion length.

Figure 7
Fig. 7 Simulation results of the temperature at the surface and the center of particles during cooling with initial radius of a = 2.5 µm, 10 µm, and 25 µm assuming initial temperature of 6000 K.
summarizes the simulation results showing the temperature at the center and the surface of particles having a radius of 2.5 µm, 10 µm and 25 µm. The initial temperature was set again to 6000 K to enable direct comparison with the model results. The simulation results highlight the difference in the temperature in the center and surface of the particle which increases with the size of the particle. For the 2.5 µm radius particles, the modeling and simulation results for the surface temperature of the particle are practically the same. For the 10 µm radius, the simulation results show a much slower cooling with the temperature dropping below 5000 K at about 19 µs and 56 µs on the surface and center of the particle,respectively. Assuming a speed of 100 m/s for these particles, the time to cover the distance from fiber 5 to fiber 15 would be about 30 µs. During this time, the corresponding results shown in Fig. 7 suggest that the temperature would decrease to about 4800 K on the surface of the particle and 5600 K at the center. For faster particles the changes would be even smaller. Turning our attention to the largest particles represented in the simulation by the 25 µm radius particles, the results shown in Fig. 7 suggest that the temperature at the center would only change by a few degrees during the initial 30 µs while the surface temperature would decline to about 5200 K. Assuming emission of these largest particles is thermal in nature, the recorded emission would be dominated by the hottest part of the particle which is located in the central region. In this case, the measured spectrum from these particles will remain largely unchanged as these particles travel through the interrogation volume of all collection fibers. This behavior is observed in the emission spectra collected from fiber 10 through 15 (distance of about 2.8 mm to 4.2 mm from the surface) suggesting that the temperature at the center of the largest particles is on the order of 5200 K.

4. Discussion

The emission recorded in the 400 nm to 750 nm spectral range obtained from each fiber represents the time- integrated contribution of the plasma and the thermal emission from the ejected particles. The signal collected from fibers 1 through 4 may be assigned to two main components: i) the emission of the plasma and ii) the thermal emission from the hot particles of groups 1 and 2 which were produced by localized heating of the material by the laser pulse and travel through this distance in the presence of a high (on the order of 1 eV or higher) plasma temperature. The spectral profile of each component may not significantly change for fibers 1 through 4 (although heating of the smaller particles by the plasma is possible) but their relative intensity will change as the plasma electron density is reduced with distance. On the other hand, the group 3 particles consist mostly of cold material while the thermal emission of the hotter particles will continuously red-shift as they traverse (at low speed) through the interrogation volume near the surface and remain outside the spectral range recorded in this experiment. Consequently, the apparent blue-shifting of the integrated spectrum observed from fibers 1 through 4 can be attributed to the complex characteristics of the time-integrated behavior of the plume in this region and will not be further investigated in this study.

There are also a number of secondary mechanisms that may be important for a more detailed interpretation of the experimental results. For example, the plasma temperature and density is declining as a function of distance in a continuous way, so there is no well-defined plasma region. In addition, we have experimentally observed using our system (see Fig. 2(b)) that the plasma outline varies from shot to shot, presumably affected by laser energy fluctuations and amount of the material volume that is involved in the breakdown process. As the spectra were integrated over a large number of events, this may be reflected on the loss of definition on the exact distance from the surface (fiber number) where the particles exhibit the rapid cooling phase. Furthermore, the evaporative cooling process of particles (such as the particles in group 2) may be reduced by the buildup of the Si and O vapor pressure arising from the evaporation of the faster and smaller particles. The simple model employed in this work assumes an ideal blackbody emission by the particles, which may not be correct as the emissivity of the particles is expected to be a function of their size, temperature, wavelength, and chemistry. During the period of time that the particles are cooling down via evaporation, their size is also changing. In addition, Kirchhoff's law of thermal radiation relates directly the emissivity with the absorptivity. The absorptivity of fused silica changes with temperature. Specifically, the absorption spectrum of silica red-shifts from ultraviolet to the near ultraviolet and further to the visible range with increasing temperature [32

32. B. Sadigh, P. Erhart, D. Åberg, A. Trave, E. Schwegler, and J. Bude, “First-principles calculations of the Urbach tail in the optical absorption spectra of silica glass,” Phys. Rev. Lett. 106(2), 027401 (2011). [CrossRef] [PubMed]

]. This phenomenon leads to two counteracting effects that can influence the relationship between the observed spectral peak and corresponding particle temperature. First, the blue-shift in the emissivity as the particles are cooling down will cause a corresponding blue-shift in the thermal emission spectrum. As the emission spectrum is proportional to the emissivity and blackbody functions (for simplicity we approximate that the absorptivity is independent of wavelength over the visible range), this effect may lead to a thermal emission spectral peak that corresponds to a higher temperature than that predicted by the blackbody spectrum. Second, the emission of the hotter central region of a larger particle, which dominates the recorded emission of the more distant fibers, passes through and is filtered by the outer, near-surface region of the particle (that is of lower temperature (see Fig. 7)). As the absorptivity of the outer region near the surface is higher at shorter wavelengths, the detected emission from the center hotter region will be red-shifted corresponding to a lower estimated temperature. It is not clear how these two counteracting mechanism affect the position of the peak of the spectrum. However, we can assume that the absorptivity at the temperatures involved in this work does not rapidly change near the peak of the thermal emission, and therefore its influence of the position of the peak of the emission may be relatively small.

This work suggests that the initial temperature of the ejected particles is on the order of 0.5 eV or higher. This is in agreement with the temperatures observed at the initial phase of a nanosecond breakdown event (observed in the bulk) in fused silica which also indicated that the material temperature at early times is on the order of 1 eV [33

33. C. W. Carr, H. B. Radousky, A. M. Rubenchik, M. D. Feit, and S. G. Demos, “Localized dynamics during laser-induced damage in optical materials,” Phys. Rev. Lett. 92(8), 087401 (2004). [CrossRef] [PubMed]

]. The integrated light-scattering images presented in this work (Fig. 2(c)) demonstrate that there is a group of particles that change direction of propagation which often leads to a complete turnaround and impact on the surface after having traveled hundreds of microns from it. This effect may be responsible for contamination of the surrounding surface with debris following localized damage of the optical element.

5. Conclusion

Understanding the physical properties of the ejecta, namely the kinetic and thermodynamic properties, can offer valuable insight into the dynamics of their formation and allow for more accurate predictions of the characteristics of the final damage crater and amount of material removed. In addition, knowledge on the spatio-temporal distribution of any material re-deposited on the surface is valuable for a number of applications such as laser-induced damage, laser ablation, and laser-assisted thin film deposition. This work provides new information on these processes, including information on the initial temperature of the particles, the cooling process, and the presence of pressure gradient-driven (or other type of) forces acting on the particles.

Acknowledgments

We thank Dr. Christopher J. Stolz and Dr. Frank Wagner for assistance and stimulating discussions. This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

References and links

1.

A. K. Burnham, L. Hackel, P. Wegner, T. Parham, L. Hrubesh, B. Penetrante, P. Whitman, S. Demos, J. Menapace, M. Runkel, M. Fluss, M. Feit, M. Key, and T. Biesiada, “Improving 351-nm damage performance of large-aperture fused silica and DKDP optics,” Proc. SPIE 4679, 173–185 (2002). [CrossRef]

2.

S. S. Harilal, C. V. Bindhu, R. C. Issac, V. P. N. Nampoori, and C. P. G. Vallabhan, “Electron density and temperature measurements in a laser produced carbon plasma,” J. Appl. Phys. 82(5), 2140–2146 (1997). [CrossRef]

3.

J. Hermann, C. Boulmer-Leborgne, and D. Hong, “Diagnostics of the early phase of an ultraviolet laser induced plasma by spectral line analysis considering self-absorption,” J. Appl. Phys. 83(2), 691–696 (1998). [CrossRef]

4.

H. C. Liu, X. L. Mao, J. H. Yoo, and R. E. Russo, “Early phase laser induced plasma diagnostics and mass removal during single-pulse laser ablation of silicon,” Spectrochim. Acta B 54(11), 1607–1624 (1999). [CrossRef]

5.

M. Milán and J. J. Laserna, “Diagnostics of silicon plasmas produced by visible nanosecond laser ablation,” Spectrochim. Acta B 56(3), 275–288 (2001). [CrossRef]

6.

M. A. Hafez, M. A. Khedr, F. F. Elaksher, and Y. E. Gamal, “Characteristics of Cu plasma produced by a laser interaction with a solid target,” Plasma Sources Sci. Technol. 12(2), 185–198 (2003). [CrossRef]

7.

C. Aragón and J. A. Aguilera, “Characterization of laser induced plasmas by optical emission spectroscopy: a review of experiments and methods,” Spectrochim. Acta B 63(9), 893–916 (2008). [CrossRef]

8.

J. P. Singh and S. N. Thakur, eds., in Laser-Induced Breakdown Spectroscopy (Elsevier, Oxford, 2007).

9.

P. S. Dalyander, I. B. Gornushkin, and D. W. Hahn, “Numerical simulation of laser-induced breakdown spectroscopy: modeling of aerosol analysis with finite diffusion and vaporization effects,” Spectrochim. Acta B 63(2), 293–304 (2008). [CrossRef]

10.

J. E. Carranza and D. W. Hahn, “Assessment of the upper particle size limit for quantitative analysis of aerosols using laser-induced breakdown spectroscopy,” Anal. Chem. 74(21), 5450–5454 (2002). [CrossRef] [PubMed]

11.

V. Hohreiter and D. W. Hahn, “Plasma-particle interactions in a laser-induced plasma: implications for laser-induced breakdown spectroscopy,” Anal. Chem. 78(5), 1509–1514 (2006). [CrossRef] [PubMed]

12.

G. M. Hieftje, R. M. Miller, Y. Pak, and E. P. Wittig, “Theoretical examination of solute particle vaporization in analytical atomic spectrometry,” Anal. Chem. 59(24), 2861–2872 (1987). [CrossRef]

13.

G. A. Lithgow and S. G. Buckley, “Influence of particle location within plasma and focal volume on precision of single-particle laser-induced breakdown spectroscopy measurements,” Spectrochim. Acta B 60(7-8), 1060–1069 (2005). [CrossRef]

14.

I. B. Gornushkin, A. Ya. Kazakov, N. Omenetto, B. W. Smith, and J. D. Winefordner, “Radiation dynamics of post-breakdown laser induced plasma,” Spectrochim. Acta B 59(4), 401–418 (2004). [CrossRef]

15.

S. Amoruso, R. Bruzzese, N. Spinelli, R. Velotta, M. Vitiello, X. Wang, G. Ausanio, V. Iannotti, and L. Lanotte, “Generation of silicon nanoparticles via femtosecond laser ablation in vacuum,” Appl. Phys. Lett. 84(22), 4502–4505 (2004). [CrossRef]

16.

R. N. Raman, R. A. Negres, and S. G. Demos, “Kinetics of ejected particles during breakdown in fused silica by nanosecond laser pulses,” Appl. Phys. Lett. 98(5), 051901 (2011). [CrossRef]

17.

R. N. Raman, R. A. Negres, and S. G. Demos, “Time-resolved microscope system to image material response following localized laser energy deposition: exit surface damage in fused silica as a case example,” Opt. Eng. 50(1), 013602 (2011). [CrossRef]

18.

G. Koren and U. P. Oppenheim, “Laser ablation of polymers in pressurized gas ambients,” Appl. Phys. B 42(1), 41–43 (1987). [CrossRef]

19.

A. Miotello, R. Kelly, B. Braren, and C. E. Otis, “Novel geometric effects observed in debris when polymers are laser sputtered,” Appl. Phys. Lett. 61(23), 2784–2786 (1992). [CrossRef]

20.

F. Wagner and P. Hoffmann, “Structure formation in excimer laser ablation of stretched poly(ethylene therepthalate) (PET): the influence of scanning ablation,” Appl. Phys., A Mater. Sci. Process. 69(7), S841–S844 (1999). [CrossRef]

21.

J. T. Dickinson, S. C. Langford, J. J. Shin, and D. L. Doering, “Positive ion emission from excimer laser excited MgO surfaces,” Phys. Rev. Lett. 73(19), 2630–2633 (1994). [CrossRef] [PubMed]

22.

S. R. George, J. A. Leraas, S. C. Langford, and J. T. Dickinson, “Interaction of vacuum ultraviolet excimer laser radiation with fused silica. I. Positive ion emission,” J. Appl. Phys. 107(3), 033107 (2010). [CrossRef]

23.

V. Narayanan, V. Singh, P. K. Pandey, N. Shukla, and R. K. Thareja, “Increasing lifetime of the plasma channel formed in air using picoseconds and nanosecond laser pulses,” J. Appl. Phys. 101(7), 073301 (2007). [CrossRef]

24.

A. Huber, I. Beigman, D. Borodin, P. Mertens, V. Philipps, A. Pospieszczyk, U. Samm, B. Schweer, G. Sergienko, and L. Vainshtein, “Spectroscopic observation of Si I- and Si II- emission lines in the boundary of TEXTOR and comparison with kinetic calculations,” Plasma Phys. Contr. Fusion 45(2), 89–103 (2003). [CrossRef]

25.

S. Elhadj, M. J. Matthews, S. T. Yang, and D. J. Cooke, “Evaporation kinetics of laser heated silica in reactive and inert gases based on near-equilibrium dynamics,” Opt. Express 20(2), 1575–1587 (2012). [CrossRef] [PubMed]

26.

S. I. Anisimov and V. A. Khokhlov, Instabilities in Laser-Matter Interaction (CRC Press, 1999).

27.

H. L. Schick, “Thermodynamic analysis of the high temperature vaporization properties of silica,” Chem. Rev. 60(4), 331–362 (1960). [CrossRef]

28.

S. Elhadj, S. R. Qiu, A. M. Monterrosa, and C. J. Stolz, “Heating dynamics of CO2-laser irradiated silica particles with evaporative shrinking: measurements and modeling,” J. Appl. Phys. 111(9), 093113 (2012). [CrossRef]

29.

R. Brückner, “Properties and structure of vitreous silica. I,” J. Non-Cryst. Solids 5(2), 123–175 (1970). [CrossRef]

30.

S. T. Yang, M. J. Matthews, S. Elhadj, V. G. Draggoo, and S. E. Bisson, “Thermal transport in CO2 laser irradiated fused silica: in situ measurements and analysis,” J. Appl. Phys. 106(10), 103106 (2009). [CrossRef]

31.

F. Armero and J. C. Simo, “A new unconditionally stable fractional step method for nonlinear coupled thermomechanical problems,” Int. J. Numer. Methods Eng. 35(4), 737–766 (1992). [CrossRef]

32.

B. Sadigh, P. Erhart, D. Åberg, A. Trave, E. Schwegler, and J. Bude, “First-principles calculations of the Urbach tail in the optical absorption spectra of silica glass,” Phys. Rev. Lett. 106(2), 027401 (2011). [CrossRef] [PubMed]

33.

C. W. Carr, H. B. Radousky, A. M. Rubenchik, M. D. Feit, and S. G. Demos, “Localized dynamics during laser-induced damage in optical materials,” Phys. Rev. Lett. 92(8), 087401 (2004). [CrossRef] [PubMed]

OCIS Codes
(110.4190) Imaging systems : Multiple imaging
(140.3440) Lasers and laser optics : Laser-induced breakdown
(300.2140) Spectroscopy : Emission
(350.4990) Other areas of optics : Particles
(110.6915) Imaging systems : Time imaging

ToC Category:
Laser Microfabrication

History
Original Manuscript: August 13, 2012
Manuscript Accepted: October 24, 2012
Published: November 29, 2012

Citation
Rajesh N. Raman, Selim Elhadj, Raluca A. Negres, Manyalibo J. Matthews, Michael D. Feit, and Stavros G. Demos, "Characterization of ejected fused silica particles following surface breakdown with nanosecond pulses," Opt. Express 20, 27708-27724 (2012)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-20-25-27708


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. A. K. Burnham, L. Hackel, P. Wegner, T. Parham, L. Hrubesh, B. Penetrante, P. Whitman, S. Demos, J. Menapace, M. Runkel, M. Fluss, M. Feit, M. Key, and T. Biesiada, “Improving 351-nm damage performance of large-aperture fused silica and DKDP optics,” Proc. SPIE4679, 173–185 (2002). [CrossRef]
  2. S. S. Harilal, C. V. Bindhu, R. C. Issac, V. P. N. Nampoori, and C. P. G. Vallabhan, “Electron density and temperature measurements in a laser produced carbon plasma,” J. Appl. Phys.82(5), 2140–2146 (1997). [CrossRef]
  3. J. Hermann, C. Boulmer-Leborgne, and D. Hong, “Diagnostics of the early phase of an ultraviolet laser induced plasma by spectral line analysis considering self-absorption,” J. Appl. Phys.83(2), 691–696 (1998). [CrossRef]
  4. H. C. Liu, X. L. Mao, J. H. Yoo, and R. E. Russo, “Early phase laser induced plasma diagnostics and mass removal during single-pulse laser ablation of silicon,” Spectrochim. Acta B54(11), 1607–1624 (1999). [CrossRef]
  5. M. Milán and J. J. Laserna, “Diagnostics of silicon plasmas produced by visible nanosecond laser ablation,” Spectrochim. Acta B56(3), 275–288 (2001). [CrossRef]
  6. M. A. Hafez, M. A. Khedr, F. F. Elaksher, and Y. E. Gamal, “Characteristics of Cu plasma produced by a laser interaction with a solid target,” Plasma Sources Sci. Technol.12(2), 185–198 (2003). [CrossRef]
  7. C. Aragón and J. A. Aguilera, “Characterization of laser induced plasmas by optical emission spectroscopy: a review of experiments and methods,” Spectrochim. Acta B63(9), 893–916 (2008). [CrossRef]
  8. J. P. Singh and S. N. Thakur, eds., in Laser-Induced Breakdown Spectroscopy (Elsevier, Oxford, 2007).
  9. P. S. Dalyander, I. B. Gornushkin, and D. W. Hahn, “Numerical simulation of laser-induced breakdown spectroscopy: modeling of aerosol analysis with finite diffusion and vaporization effects,” Spectrochim. Acta B63(2), 293–304 (2008). [CrossRef]
  10. J. E. Carranza and D. W. Hahn, “Assessment of the upper particle size limit for quantitative analysis of aerosols using laser-induced breakdown spectroscopy,” Anal. Chem.74(21), 5450–5454 (2002). [CrossRef] [PubMed]
  11. V. Hohreiter and D. W. Hahn, “Plasma-particle interactions in a laser-induced plasma: implications for laser-induced breakdown spectroscopy,” Anal. Chem.78(5), 1509–1514 (2006). [CrossRef] [PubMed]
  12. G. M. Hieftje, R. M. Miller, Y. Pak, and E. P. Wittig, “Theoretical examination of solute particle vaporization in analytical atomic spectrometry,” Anal. Chem.59(24), 2861–2872 (1987). [CrossRef]
  13. G. A. Lithgow and S. G. Buckley, “Influence of particle location within plasma and focal volume on precision of single-particle laser-induced breakdown spectroscopy measurements,” Spectrochim. Acta B60(7-8), 1060–1069 (2005). [CrossRef]
  14. I. B. Gornushkin, A. Ya. Kazakov, N. Omenetto, B. W. Smith, and J. D. Winefordner, “Radiation dynamics of post-breakdown laser induced plasma,” Spectrochim. Acta B59(4), 401–418 (2004). [CrossRef]
  15. S. Amoruso, R. Bruzzese, N. Spinelli, R. Velotta, M. Vitiello, X. Wang, G. Ausanio, V. Iannotti, and L. Lanotte, “Generation of silicon nanoparticles via femtosecond laser ablation in vacuum,” Appl. Phys. Lett.84(22), 4502–4505 (2004). [CrossRef]
  16. R. N. Raman, R. A. Negres, and S. G. Demos, “Kinetics of ejected particles during breakdown in fused silica by nanosecond laser pulses,” Appl. Phys. Lett.98(5), 051901 (2011). [CrossRef]
  17. R. N. Raman, R. A. Negres, and S. G. Demos, “Time-resolved microscope system to image material response following localized laser energy deposition: exit surface damage in fused silica as a case example,” Opt. Eng.50(1), 013602 (2011). [CrossRef]
  18. G. Koren and U. P. Oppenheim, “Laser ablation of polymers in pressurized gas ambients,” Appl. Phys. B42(1), 41–43 (1987). [CrossRef]
  19. A. Miotello, R. Kelly, B. Braren, and C. E. Otis, “Novel geometric effects observed in debris when polymers are laser sputtered,” Appl. Phys. Lett.61(23), 2784–2786 (1992). [CrossRef]
  20. F. Wagner and P. Hoffmann, “Structure formation in excimer laser ablation of stretched poly(ethylene therepthalate) (PET): the influence of scanning ablation,” Appl. Phys., A Mater. Sci. Process.69(7), S841–S844 (1999). [CrossRef]
  21. J. T. Dickinson, S. C. Langford, J. J. Shin, and D. L. Doering, “Positive ion emission from excimer laser excited MgO surfaces,” Phys. Rev. Lett.73(19), 2630–2633 (1994). [CrossRef] [PubMed]
  22. S. R. George, J. A. Leraas, S. C. Langford, and J. T. Dickinson, “Interaction of vacuum ultraviolet excimer laser radiation with fused silica. I. Positive ion emission,” J. Appl. Phys.107(3), 033107 (2010). [CrossRef]
  23. V. Narayanan, V. Singh, P. K. Pandey, N. Shukla, and R. K. Thareja, “Increasing lifetime of the plasma channel formed in air using picoseconds and nanosecond laser pulses,” J. Appl. Phys.101(7), 073301 (2007). [CrossRef]
  24. A. Huber, I. Beigman, D. Borodin, P. Mertens, V. Philipps, A. Pospieszczyk, U. Samm, B. Schweer, G. Sergienko, and L. Vainshtein, “Spectroscopic observation of Si I- and Si II- emission lines in the boundary of TEXTOR and comparison with kinetic calculations,” Plasma Phys. Contr. Fusion45(2), 89–103 (2003). [CrossRef]
  25. S. Elhadj, M. J. Matthews, S. T. Yang, and D. J. Cooke, “Evaporation kinetics of laser heated silica in reactive and inert gases based on near-equilibrium dynamics,” Opt. Express20(2), 1575–1587 (2012). [CrossRef] [PubMed]
  26. S. I. Anisimov and V. A. Khokhlov, Instabilities in Laser-Matter Interaction (CRC Press, 1999).
  27. H. L. Schick, “Thermodynamic analysis of the high temperature vaporization properties of silica,” Chem. Rev.60(4), 331–362 (1960). [CrossRef]
  28. S. Elhadj, S. R. Qiu, A. M. Monterrosa, and C. J. Stolz, “Heating dynamics of CO2-laser irradiated silica particles with evaporative shrinking: measurements and modeling,” J. Appl. Phys.111(9), 093113 (2012). [CrossRef]
  29. R. Brückner, “Properties and structure of vitreous silica. I,” J. Non-Cryst. Solids5(2), 123–175 (1970). [CrossRef]
  30. S. T. Yang, M. J. Matthews, S. Elhadj, V. G. Draggoo, and S. E. Bisson, “Thermal transport in CO2 laser irradiated fused silica: in situ measurements and analysis,” J. Appl. Phys.106(10), 103106 (2009). [CrossRef]
  31. F. Armero and J. C. Simo, “A new unconditionally stable fractional step method for nonlinear coupled thermomechanical problems,” Int. J. Numer. Methods Eng.35(4), 737–766 (1992). [CrossRef]
  32. B. Sadigh, P. Erhart, D. Åberg, A. Trave, E. Schwegler, and J. Bude, “First-principles calculations of the Urbach tail in the optical absorption spectra of silica glass,” Phys. Rev. Lett.106(2), 027401 (2011). [CrossRef] [PubMed]
  33. C. W. Carr, H. B. Radousky, A. M. Rubenchik, M. D. Feit, and S. G. Demos, “Localized dynamics during laser-induced damage in optical materials,” Phys. Rev. Lett.92(8), 087401 (2004). [CrossRef] [PubMed]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited