- Email: [email protected]
New damage behavior induced by nanosecond laser pulses on the surface of silica films
Optics & Laser Technology 46 (2013) 77–80
Contents lists available at SciVerse ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
New damage behavior induced by nanosecond laser pulses on the surface of silica films Zhilin Xia a,b,n, Dawei Li c, Yuan’an Zhao c, Yuting Wu a a
School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070, China State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, China c Shanghai Institution of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China b
a r t i c l e i n f o
abstract
Article history: Received 20 February 2012 Received in revised form 23 April 2012 Accepted 23 April 2012 Available online 28 May 2012
The discovery of subwavelength periodic stripes with similar grating structures is currently only associated with femto-/picosecond laser ablation experiments. A nanosecond laser is generally accepted as incapable of etching out subwavelength periodic stripes. While in this paper, the subwavelength periodic stripes on the surface of silica films induced by nanosecond laser pulses have been observed. The silica films have a particle accumulation structure. This laser damage behavior promotes the industrial production advancement of laser etching subwavelength periodic stripes. The microstructure and spatial period of the periodic stripes under different film structures and radiation energy densities were studied. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Subwavelength periodic stripe Nanosecond laser etching Silica films
1. Introduction Nanofabrication techniques have resulted in widespread interest on the novel electronic, optical, and mechanical properties of various nanostructures. Among the physical, chemical, and mechanical nanofabrication techniques, laser-assisted methods are efficient and environmentally friendly [1]. In laser ablation, the remarkable formation of laser-induced periodic surface structures occurs in various materials under diverse irradiation conditions [2]. It can be seen from an earlier study on laser ablation that laser irradiation can facilitate the formation of periodic structures on a material surface [3]. Before the advent of the femtosecond laser, a long-pulse or continuous laser was used to obtain stripes with large periods. In a normal-incident situation, the intervals of the stripes are generally close to the laser wavelength. These are called as ‘‘classic stripes’’ [4]. In contrast, ultrashort pulse laserinduced novel stripes have spatial periods considerably less than the laser wavelength [5]. These stripes are strictly perpendicular to the laser polarization direction. Their structures are similar with those of grating, wherein a steep edge exists between the slot and ridge parts. This kind of ‘‘subwavelength periodic stripes’’ has opened a new avenue for laser micro–nano production.
n Corresponding author at: Wuhan University of Technology, School of Materials Science and Engineering, Luoshi road 122#, Wuhan 430070, China. Tel./fax: þ86 27 87863563. E-mail address: [email protected] (Z. Xia).
0030-3992/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.optlastec.2012.04.032
Consequently, subwavelength periodic stripes have become an important topic in laser studies and related fields. The femtosecond laser-etching process of subwavelength periodic stripes has been studied for over 10 years. Numerous theories, such as those on scattering waves, organization, secondary harmonics, as well as laser and plasma interactions, have been proposed [6–10]. Huang et al. [10] have systematically researched the laser and plasma interaction theory. The surface plasma interference-formed standing wave is shown to determine the initial spatial period of subwavelength stripes. The main etching mechanism is a nanoscale Coulomb explosion in the grooves of the stripes. The subwavelength stripe orientation depends on the cavity mode and surface plasma transverse mode (TM) wave characteristic. Thus far, only femtosecond and picosecond laser experiments have focused on subwavelength periodic stripes [11]. In the case of nanosecond laser irradiation [12–15], damage is mainly induced by the strong, random absorption of impurities or defects in a material. There is no clear pulse fluence boundary between material damage and non-damage, which limits the development and application of the nanosecond laser microprocessing technology. The heat action process is the main distinction between nanosecond and femtosecond laser damages. The local areas of the material may reach a sufficiently high pressure and erupt. The eruption process takes away the periodic stripes that the surface plasma has etched. Therefore, a nanosecond laser is considered not suitable for etching subwavelength periodic structure stripes [8]. While, plasma flash is generally observed in the nanosecond laser-induced material damage process. The core part of the
78
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
subwavelength stripe etching mechanism that Huang et al. [10] have proposed does not repel a nanosecond laser if some unfavorable factors of the damage process can be overcome. Such factors include the local area absorption and high-temperature material eruption. Hence, it is possible to use a nanosecond laser to etch subwavelength periodic stripes.
2. Experiment details 2.1. Experimental design In the present study, an experiment using nanosecond laser to irradiate sol–gel porous silicon dioxide films with a particle accumulation structure was designed based on the following observations. First, the sol–gel method can considerably improve film purity and decrease the probability of film damage. Second, visible or infrared light cannot directly instigate the production of a material surface plasma wave, but the particle structure of a sol–gel silica films can scatter incident light. The incident laser can then be aided to stimulate a surface plasma wave. Third, when porous materials melt or gasify, they can expand and penetrate to some extent. This process can reduce the pressure in a high-temperature area, and delay or prevent local hightemperature material eruption. The local temperature of the material can also be decreased to the extent that the material can be restored to its solid form, and the laser etching of the periodic stripe structure can be maintained. 2.2. Preparation of silica films Based on a 1.0: 2.0: R: 37 ratio of tetraethylorthosilicate (TEOS), H2O, NH3, and C2H5OH (R¼0.1, 0.2, 0.4, 0.6, and 0.8), the reagents were combined, completely sealed, and fully mixed at 50 1C by stirring 4–5 h. The solution was then filtered at a low pressure using a filtration membrane with a pore radius of 0.8 mm. At 20–25 1C and relative humidity of 60%, the pulling method was used to prepare the coatings. The cleaned substrate (B270 glass with the dimension of 25 mm 75 mm 2 mm) was dipped in the solution for 60 s before being smoothly, uniformly, and vertically lifted at a speed of 1.5 mm/s. The heat treatment was performed in a furnace in the following scheme: from room temperature to 100 1C at approximately 2–5 1C/min and maintained for 30 min under 100 1C; from 100 to 250 1C at approximately 2–5 1C/min and maintained for 1 h at 250 1C; as well as from 250 to 450 1C at approximately 2–5 1C/min and maintained for 0.5 h at 450 1C. Subsequently, the furnace was closed and the samples were allowed to cool to room temperature. Before annealing, a beaker filled with ammonia was placed in the muffle furnace to allow the film to be heated under an ammonia atmosphere. The final film thickness was detected by a step profiler and it is about 200 nm. 2.3. Laser etching The laser etching experiment was performed in the ‘‘1-on-1’’ regime of laser damage testing according to International Organization for Standardization 11254-1.2. The ‘‘1-on-1’’ regime means only one laser pulse will be imposed on a test point on the specimen surface. The schematic of the experimental setup for laser etching is shown in Fig. 1. The Nd: YAG laser system was operated at the TEM00 mode. The laser beam was focused on the target plane normally with 1 mm diameter spot (1/e2) by a nonspherical lens of 250 mm focal length. The laser pulse width was 12 ns, and the laser wavelength was 1064 nm. A total of 100 sites were tested for each sample and every site was exposed to
Fig. 1. The schematic of the experimental setup for laser etching Ref. [16].
one laser pulse. Every ten sites, in a line, were exposed to the same fluency, and the portion of the damaged sites was recorded. A Nomarski microscope was used to determine the damage to the radiation sites at 100 magnification. The laser pulse energy was adjusted by an attenuator comprising a half-wavelength plate and a polarizer. The polarization direction of the irradiation light was determined by the polarizer, and it was independent on the laser pulse energy. Before testing, a marking line which was parallel to the polarization direction of the irradiation light was made on the specimen. This marking line is helpful for judging the relationship between the polarization direction of the irradiation light and the orientation of the formed subwavelength ripples.
3. Results and discussions 3.1. Films structure and laser damage threshold Fig. 2 shows the scanning electron microscopy (SEM) image of the SiO2 film when R¼0.8. It can be observed that the SiO2 film has particle accumulation structure and numerous cracks exist in it. Based on the cracking trend, the cracks are deduced to be caused by the preparation process and not by the sample preparation for the SEM test, which could have affected the integrity of the etching stripes. Besides, when NH3/TEOS¼0.1, 0.2, 0.4, 0.6, and 0.8, the radii of SiO2 particles are 5, 7.5, 12.5, 20, and 30 nm, respectively. The particle radii are the average value of more than twenty measurement particles radii from the SEM images such as Fig. 2. The laser damage thresholds have been fitted according to the damage probabilities under different irradiation energy densities [16]. The damage threshold is corresponding to the highest pulse fluence which will cause zero probability damage. When the radii of SiO2 particles are 5, 7.5, 12.5, 20, and 30 nm, the laser damage thresholds are 11.2, 11.7, 16.8, 14.3, and 13 J/cm2, respectively. Compared with the silicon oxide film prepared by electron beam evaporation, the sol–gel film has higher laser damage threshold. The higher purity of the sol–gel film can account for this finding. The sol–gel film is also a loose microstructure that can ease local thermal stress caused by laser irradiation [17–18]. 3.2. Structure of periodic stripes In Fig. 3, the periodic stripes in the sol–gel films damage spots are shown as they emerge. The spatial periods of the periodic stripes in every picture are less than the laser wavelength of 1064 nm. The orientations of the stripes in a picture are the same and they are vertical to the polarization of the laser irradiation.
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
79
Besides, these stripes structures have a similarity to nanograting structures, with a steep edge between the slot and ridge. These stripes show basic characteristics that only appear in subwavelength periodic stripes etched by a femtosecond laser, and do not belong to classic stripes. These subwavelength periodic stripes can be observed only in the damage pit. Neither the slot nor the ridge reserves its original particle accumulation structure. This indicates that the stripe structure is formed after the materials melting. It can be seen by comparing fig. 3(a)–(d) that higher pulse fluence results in a better integrity of the stripe and a larger
duty cycle of the stripe structure. However, as Fig. 4 shows, excess pulse fluence leads to material eruption and inability to sustain the etching of stripes. Stripe spatial period measurements were conducted in at least three different areas of the damage spot. The average spatial period of the same energy density laser irradiation induced was calculated. Fig. 5 shows the relationship between the subwavelength stripe spatial period and etching pulse fluence. A wide range of pulse fluence is suitable in forming subwavelength periodic stripes. The stripe spatial period is related to the pulse
Fig. 2. SEM image of the SiO2 film when R ¼0.8.
Fig. 4. SEM image of the damage spot of the SiO2 film when R ¼0.8 and pulse fluence is 47.9 J/cm2.
Fig. 3. Periodic stripes in the damage spots of sol–gel films under different R-values. Nd: YAG laser pulses with different energy densities, wavelength of 1064 nm and pulse width of 12 ns were used. (a) R¼ 0.1; pulse fluence is 19.4 J/cm2, (b) R ¼0.2; pulse fluence is 17.6 J/cm2, (c) R¼ 0.6; pulse fluence is 15.6 J/cm2, and (d) R ¼0.8; pulse fluence is 15.6 J/cm2.
80
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
the pulse fluence. Higher pulse fluence results in a larger stripe spatial period. However, the relationship between the stripe spatial period and particles sizes of the film microstructure is not clear.
Acknowledgments The authors gratefully acknowledge the financial support of the National Science Foundation of China (Grant nos. 10974150 and 10804090), the Open Research Fund of Key Laboratory of Material for High Power Lasers—Chinese Academy of Science the China Postdoctoral Science Foundation (Grant no. 2012M511691), the Self-determined and Innovative Research Funds of WUT (Grant no. 2012-IV-014). References Fig. 5. Relationship between the subwavelength stripe spatial period and etching pulse fluence. Nd: YAG laser pulses with wavelength of 1064 nm and pulse width of 12 ns were used.
fluence. Higher pulse fluence results in a larger stripe spatial period. However, the relationship between the stripe spatial period and the particle size of the sol–gel film microstructure is not clear. Plasma flash is generally observed in the ns pulsed laserinduced material damage process. So we deduced that the Huang suggested theory may be effective in explaining ns-pulsed laser induced subwavelength ripples. Based on the ns-pulsed laser induced damage mechanisms [19,20], considering the Huang suggested theory, the general process of the nanosecond laser etching of subwavelength stripes may be the case as follows: laser irradiation causes the high ionization of a material [21], produces large amounts of free electrons, and makes medium materials exhibit metallic characteristics [22]. The material temperature also rises and results in the formation of a high-pressure melt. The melted material expands to release pressure, and the temperature falls to a certain degree for a porous structure. This phenomenon prevents eruption and facilitates material solidification. The particle structure also excites high-frequency waves, which leads to the formation of a surface plasma with a TM wave feature. This wave is coupled with a laser TM wave [23,24]. A Coulomb explosion etches out grooves oriented in the vertical polarization direction of the laser radiation. In the TM-oriented grooves, the TM wave continuously strengthens, which further etches the channel.
4. Conclusions In conclusion, the difficulties of etching subwavelength stripes using a nanosecond laser pulse have been analyzed. Based on these analyses, an experiment on the nanosecond laser etching of subwavelength stripes has been designed and conducted. The sol– gel method was used to process the optical glass surface. Nanoporous-structured SiO2 films were formed on the surface of the optical glass. A Nd: YAG laser with a wavelength of 1064 nm and pulse width of 12 ns was used to etch subwavelength periodic stripes on the coatings. Nanosecond laser-etched subwavelength stripes only appear in the damage spot after material melting. Excessively high pulse fluence causes the materials to erupt, and leads to failure in sustaining the etching process. There is a wide range of pulse fluence that can form subwavelength periodic stripes. The spatial period of these stripes is related to
[1] Perrie re J, Millon E´, Fogarassy E´. Recent advances in laser processing of materials. Amsterdam: Elsevier; 2006. [2] Rajeev PP, Gertsvolf M, Simova E, Hnatovsky C, Taylor RS, Bhardwaj VR, Rayner DM, Corkum PB. Memory in nonlinear ionization of transparent solids. Physical Review Letters 2006;97:253001. [3] van Driel HM, Sipe JE, Young JF. Laser-induced periodic surface structure on solids: a universal phenomenon. Physical Review Letters 1982;49:1955. [4] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Large area uniform nanostructures fabricated by direct femtosecond laser ablation. Optics Express 2008;16: 19354. [5] Huang M. A study on formation of subwavelength periodic structures induced by ultrashort laser pulses. Doctoral dissertation: Sun Yat-sen University; 2009. [6] Mora P. Plasma expansion into a vacuum. Physical Review Letters 2003;90: 185002. [7] Sipe JE, Young JF, Preston JS, van Driel HM. Laser-induced periodic surface structure. I. theory. Physical Review B 1983;27:1141. [8] Ran LL. Self-organized periodic surface structures on ZnO induced by femtosecond laser. Applied Physics A: Materials Science and Processing 2010;100:517. [9] Jia TQ. Formation of nanogratings on the surface of a ZnSe crystal irradiated by femtosecond laser pulses. Physical Review B 2005;72:125429. [10] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Mechanisms of ultrafast laserinduced deep sub wave length gratings on graphite and diamond. Physical Review B 2009;79:125436. [11] Hsu EM, Crawford THR, Tiedje HF, Haugen HK. Periodic surface structures on gallium phosphate after irradiation with 150 fs 7 ns laser pulses at 800 nm. Applied Physics Letters 2007;91:111102. [12] Xia ZL, Shao JD, Fan ZX. Statistical approach to bulk inclusion initialized damage in films. Optics Communications 2006;265:620. [13] Wei CY, Shao JD, He HB, Yi K, Fan ZX. Mechanism initiated by nano absorber for UV nanosecond-pulse-driven damage of dielectric coatings. Optics Express 2008;16:3376. [14] Zhang DP, Wang CJ, Fan P. Influence of plasma treatment on laser-induced damage characters of HfO2 thin films at 355 nm. Optics Express 2009;17: 8246. [15] Zhang DP, Shao JD, Zhang DW. Employing oxygen-plasma post treatment to improve the laser-induced damage threshold of ZrO2 films prepared by the electron-beam evaporation method. Optics Letters 2004;29:2870. [16] Zhao YA, Wang T, Zhang DW, Shao JD, Fan ZX. laser damage accumulation effects of HfO2/SiO2 antireflective coatings. Applied Surface Science 2005;245:335. [17] Liang LP, Zhang L, Sheng YG. Studies on the laser-induced damage resistance of sol–gel derived ZrO2–TiO2 composite high refractive index films. Acta Physica Sinica 2007;56:3596. [18] Grosso D. Scandia optical coatings for application at 351 nm. Thin Solid Films 2000;368:116. [19] Xia ZL, Wang H, Xu Q. The stress relief mechanism in laser irradiating on porous films. Optical Communications 2012;285:70. [20] Wei CY, Shao JD, He HB, Yi K, Fan ZX. Mechanism initiated by nanoabsorber for UV nanosecond-pulse-driven damage of dielectric coatings. Optics Express 2008;16:3376. [21] Smith AV, Do BT. Bulk and surface laser damage of silica by picosecond and nanosecond pulses at 1064 nm. Applied Optics 2008;47:4812. [22] Du D, Liu X, Korn G, Squier J, Mourou G. Laser-induced breakdown by impact ionization in SiO2 with pulse widths from 7 ns to 150 fs. Applied Physics Letters 1994;64:3071. [23] Raether H. Surface plasmons on smooth and rough surfaces and on gratings. Springer-Verlag; 1988. [24] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Origin of laser-induced nearsubwavelength ripples: interference between surface plasmons and incident laser. ACS NANO 2009;3:4062.
Contents lists available at SciVerse ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
New damage behavior induced by nanosecond laser pulses on the surface of silica films Zhilin Xia a,b,n, Dawei Li c, Yuan’an Zhao c, Yuting Wu a a
School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070, China State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, China c Shanghai Institution of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China b
a r t i c l e i n f o
abstract
Article history: Received 20 February 2012 Received in revised form 23 April 2012 Accepted 23 April 2012 Available online 28 May 2012
The discovery of subwavelength periodic stripes with similar grating structures is currently only associated with femto-/picosecond laser ablation experiments. A nanosecond laser is generally accepted as incapable of etching out subwavelength periodic stripes. While in this paper, the subwavelength periodic stripes on the surface of silica films induced by nanosecond laser pulses have been observed. The silica films have a particle accumulation structure. This laser damage behavior promotes the industrial production advancement of laser etching subwavelength periodic stripes. The microstructure and spatial period of the periodic stripes under different film structures and radiation energy densities were studied. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Subwavelength periodic stripe Nanosecond laser etching Silica films
1. Introduction Nanofabrication techniques have resulted in widespread interest on the novel electronic, optical, and mechanical properties of various nanostructures. Among the physical, chemical, and mechanical nanofabrication techniques, laser-assisted methods are efficient and environmentally friendly [1]. In laser ablation, the remarkable formation of laser-induced periodic surface structures occurs in various materials under diverse irradiation conditions [2]. It can be seen from an earlier study on laser ablation that laser irradiation can facilitate the formation of periodic structures on a material surface [3]. Before the advent of the femtosecond laser, a long-pulse or continuous laser was used to obtain stripes with large periods. In a normal-incident situation, the intervals of the stripes are generally close to the laser wavelength. These are called as ‘‘classic stripes’’ [4]. In contrast, ultrashort pulse laserinduced novel stripes have spatial periods considerably less than the laser wavelength [5]. These stripes are strictly perpendicular to the laser polarization direction. Their structures are similar with those of grating, wherein a steep edge exists between the slot and ridge parts. This kind of ‘‘subwavelength periodic stripes’’ has opened a new avenue for laser micro–nano production.
n Corresponding author at: Wuhan University of Technology, School of Materials Science and Engineering, Luoshi road 122#, Wuhan 430070, China. Tel./fax: þ86 27 87863563. E-mail address: [email protected] (Z. Xia).
0030-3992/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.optlastec.2012.04.032
Consequently, subwavelength periodic stripes have become an important topic in laser studies and related fields. The femtosecond laser-etching process of subwavelength periodic stripes has been studied for over 10 years. Numerous theories, such as those on scattering waves, organization, secondary harmonics, as well as laser and plasma interactions, have been proposed [6–10]. Huang et al. [10] have systematically researched the laser and plasma interaction theory. The surface plasma interference-formed standing wave is shown to determine the initial spatial period of subwavelength stripes. The main etching mechanism is a nanoscale Coulomb explosion in the grooves of the stripes. The subwavelength stripe orientation depends on the cavity mode and surface plasma transverse mode (TM) wave characteristic. Thus far, only femtosecond and picosecond laser experiments have focused on subwavelength periodic stripes [11]. In the case of nanosecond laser irradiation [12–15], damage is mainly induced by the strong, random absorption of impurities or defects in a material. There is no clear pulse fluence boundary between material damage and non-damage, which limits the development and application of the nanosecond laser microprocessing technology. The heat action process is the main distinction between nanosecond and femtosecond laser damages. The local areas of the material may reach a sufficiently high pressure and erupt. The eruption process takes away the periodic stripes that the surface plasma has etched. Therefore, a nanosecond laser is considered not suitable for etching subwavelength periodic structure stripes [8]. While, plasma flash is generally observed in the nanosecond laser-induced material damage process. The core part of the
78
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
subwavelength stripe etching mechanism that Huang et al. [10] have proposed does not repel a nanosecond laser if some unfavorable factors of the damage process can be overcome. Such factors include the local area absorption and high-temperature material eruption. Hence, it is possible to use a nanosecond laser to etch subwavelength periodic stripes.
2. Experiment details 2.1. Experimental design In the present study, an experiment using nanosecond laser to irradiate sol–gel porous silicon dioxide films with a particle accumulation structure was designed based on the following observations. First, the sol–gel method can considerably improve film purity and decrease the probability of film damage. Second, visible or infrared light cannot directly instigate the production of a material surface plasma wave, but the particle structure of a sol–gel silica films can scatter incident light. The incident laser can then be aided to stimulate a surface plasma wave. Third, when porous materials melt or gasify, they can expand and penetrate to some extent. This process can reduce the pressure in a high-temperature area, and delay or prevent local hightemperature material eruption. The local temperature of the material can also be decreased to the extent that the material can be restored to its solid form, and the laser etching of the periodic stripe structure can be maintained. 2.2. Preparation of silica films Based on a 1.0: 2.0: R: 37 ratio of tetraethylorthosilicate (TEOS), H2O, NH3, and C2H5OH (R¼0.1, 0.2, 0.4, 0.6, and 0.8), the reagents were combined, completely sealed, and fully mixed at 50 1C by stirring 4–5 h. The solution was then filtered at a low pressure using a filtration membrane with a pore radius of 0.8 mm. At 20–25 1C and relative humidity of 60%, the pulling method was used to prepare the coatings. The cleaned substrate (B270 glass with the dimension of 25 mm 75 mm 2 mm) was dipped in the solution for 60 s before being smoothly, uniformly, and vertically lifted at a speed of 1.5 mm/s. The heat treatment was performed in a furnace in the following scheme: from room temperature to 100 1C at approximately 2–5 1C/min and maintained for 30 min under 100 1C; from 100 to 250 1C at approximately 2–5 1C/min and maintained for 1 h at 250 1C; as well as from 250 to 450 1C at approximately 2–5 1C/min and maintained for 0.5 h at 450 1C. Subsequently, the furnace was closed and the samples were allowed to cool to room temperature. Before annealing, a beaker filled with ammonia was placed in the muffle furnace to allow the film to be heated under an ammonia atmosphere. The final film thickness was detected by a step profiler and it is about 200 nm. 2.3. Laser etching The laser etching experiment was performed in the ‘‘1-on-1’’ regime of laser damage testing according to International Organization for Standardization 11254-1.2. The ‘‘1-on-1’’ regime means only one laser pulse will be imposed on a test point on the specimen surface. The schematic of the experimental setup for laser etching is shown in Fig. 1. The Nd: YAG laser system was operated at the TEM00 mode. The laser beam was focused on the target plane normally with 1 mm diameter spot (1/e2) by a nonspherical lens of 250 mm focal length. The laser pulse width was 12 ns, and the laser wavelength was 1064 nm. A total of 100 sites were tested for each sample and every site was exposed to
Fig. 1. The schematic of the experimental setup for laser etching Ref. [16].
one laser pulse. Every ten sites, in a line, were exposed to the same fluency, and the portion of the damaged sites was recorded. A Nomarski microscope was used to determine the damage to the radiation sites at 100 magnification. The laser pulse energy was adjusted by an attenuator comprising a half-wavelength plate and a polarizer. The polarization direction of the irradiation light was determined by the polarizer, and it was independent on the laser pulse energy. Before testing, a marking line which was parallel to the polarization direction of the irradiation light was made on the specimen. This marking line is helpful for judging the relationship between the polarization direction of the irradiation light and the orientation of the formed subwavelength ripples.
3. Results and discussions 3.1. Films structure and laser damage threshold Fig. 2 shows the scanning electron microscopy (SEM) image of the SiO2 film when R¼0.8. It can be observed that the SiO2 film has particle accumulation structure and numerous cracks exist in it. Based on the cracking trend, the cracks are deduced to be caused by the preparation process and not by the sample preparation for the SEM test, which could have affected the integrity of the etching stripes. Besides, when NH3/TEOS¼0.1, 0.2, 0.4, 0.6, and 0.8, the radii of SiO2 particles are 5, 7.5, 12.5, 20, and 30 nm, respectively. The particle radii are the average value of more than twenty measurement particles radii from the SEM images such as Fig. 2. The laser damage thresholds have been fitted according to the damage probabilities under different irradiation energy densities [16]. The damage threshold is corresponding to the highest pulse fluence which will cause zero probability damage. When the radii of SiO2 particles are 5, 7.5, 12.5, 20, and 30 nm, the laser damage thresholds are 11.2, 11.7, 16.8, 14.3, and 13 J/cm2, respectively. Compared with the silicon oxide film prepared by electron beam evaporation, the sol–gel film has higher laser damage threshold. The higher purity of the sol–gel film can account for this finding. The sol–gel film is also a loose microstructure that can ease local thermal stress caused by laser irradiation [17–18]. 3.2. Structure of periodic stripes In Fig. 3, the periodic stripes in the sol–gel films damage spots are shown as they emerge. The spatial periods of the periodic stripes in every picture are less than the laser wavelength of 1064 nm. The orientations of the stripes in a picture are the same and they are vertical to the polarization of the laser irradiation.
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
79
Besides, these stripes structures have a similarity to nanograting structures, with a steep edge between the slot and ridge. These stripes show basic characteristics that only appear in subwavelength periodic stripes etched by a femtosecond laser, and do not belong to classic stripes. These subwavelength periodic stripes can be observed only in the damage pit. Neither the slot nor the ridge reserves its original particle accumulation structure. This indicates that the stripe structure is formed after the materials melting. It can be seen by comparing fig. 3(a)–(d) that higher pulse fluence results in a better integrity of the stripe and a larger
duty cycle of the stripe structure. However, as Fig. 4 shows, excess pulse fluence leads to material eruption and inability to sustain the etching of stripes. Stripe spatial period measurements were conducted in at least three different areas of the damage spot. The average spatial period of the same energy density laser irradiation induced was calculated. Fig. 5 shows the relationship between the subwavelength stripe spatial period and etching pulse fluence. A wide range of pulse fluence is suitable in forming subwavelength periodic stripes. The stripe spatial period is related to the pulse
Fig. 2. SEM image of the SiO2 film when R ¼0.8.
Fig. 4. SEM image of the damage spot of the SiO2 film when R ¼0.8 and pulse fluence is 47.9 J/cm2.
Fig. 3. Periodic stripes in the damage spots of sol–gel films under different R-values. Nd: YAG laser pulses with different energy densities, wavelength of 1064 nm and pulse width of 12 ns were used. (a) R¼ 0.1; pulse fluence is 19.4 J/cm2, (b) R ¼0.2; pulse fluence is 17.6 J/cm2, (c) R¼ 0.6; pulse fluence is 15.6 J/cm2, and (d) R ¼0.8; pulse fluence is 15.6 J/cm2.
80
Z. Xia et al. / Optics & Laser Technology 46 (2013) 77–80
the pulse fluence. Higher pulse fluence results in a larger stripe spatial period. However, the relationship between the stripe spatial period and particles sizes of the film microstructure is not clear.
Acknowledgments The authors gratefully acknowledge the financial support of the National Science Foundation of China (Grant nos. 10974150 and 10804090), the Open Research Fund of Key Laboratory of Material for High Power Lasers—Chinese Academy of Science the China Postdoctoral Science Foundation (Grant no. 2012M511691), the Self-determined and Innovative Research Funds of WUT (Grant no. 2012-IV-014). References Fig. 5. Relationship between the subwavelength stripe spatial period and etching pulse fluence. Nd: YAG laser pulses with wavelength of 1064 nm and pulse width of 12 ns were used.
fluence. Higher pulse fluence results in a larger stripe spatial period. However, the relationship between the stripe spatial period and the particle size of the sol–gel film microstructure is not clear. Plasma flash is generally observed in the ns pulsed laserinduced material damage process. So we deduced that the Huang suggested theory may be effective in explaining ns-pulsed laser induced subwavelength ripples. Based on the ns-pulsed laser induced damage mechanisms [19,20], considering the Huang suggested theory, the general process of the nanosecond laser etching of subwavelength stripes may be the case as follows: laser irradiation causes the high ionization of a material [21], produces large amounts of free electrons, and makes medium materials exhibit metallic characteristics [22]. The material temperature also rises and results in the formation of a high-pressure melt. The melted material expands to release pressure, and the temperature falls to a certain degree for a porous structure. This phenomenon prevents eruption and facilitates material solidification. The particle structure also excites high-frequency waves, which leads to the formation of a surface plasma with a TM wave feature. This wave is coupled with a laser TM wave [23,24]. A Coulomb explosion etches out grooves oriented in the vertical polarization direction of the laser radiation. In the TM-oriented grooves, the TM wave continuously strengthens, which further etches the channel.
4. Conclusions In conclusion, the difficulties of etching subwavelength stripes using a nanosecond laser pulse have been analyzed. Based on these analyses, an experiment on the nanosecond laser etching of subwavelength stripes has been designed and conducted. The sol– gel method was used to process the optical glass surface. Nanoporous-structured SiO2 films were formed on the surface of the optical glass. A Nd: YAG laser with a wavelength of 1064 nm and pulse width of 12 ns was used to etch subwavelength periodic stripes on the coatings. Nanosecond laser-etched subwavelength stripes only appear in the damage spot after material melting. Excessively high pulse fluence causes the materials to erupt, and leads to failure in sustaining the etching process. There is a wide range of pulse fluence that can form subwavelength periodic stripes. The spatial period of these stripes is related to
[1] Perrie re J, Millon E´, Fogarassy E´. Recent advances in laser processing of materials. Amsterdam: Elsevier; 2006. [2] Rajeev PP, Gertsvolf M, Simova E, Hnatovsky C, Taylor RS, Bhardwaj VR, Rayner DM, Corkum PB. Memory in nonlinear ionization of transparent solids. Physical Review Letters 2006;97:253001. [3] van Driel HM, Sipe JE, Young JF. Laser-induced periodic surface structure on solids: a universal phenomenon. Physical Review Letters 1982;49:1955. [4] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Large area uniform nanostructures fabricated by direct femtosecond laser ablation. Optics Express 2008;16: 19354. [5] Huang M. A study on formation of subwavelength periodic structures induced by ultrashort laser pulses. Doctoral dissertation: Sun Yat-sen University; 2009. [6] Mora P. Plasma expansion into a vacuum. Physical Review Letters 2003;90: 185002. [7] Sipe JE, Young JF, Preston JS, van Driel HM. Laser-induced periodic surface structure. I. theory. Physical Review B 1983;27:1141. [8] Ran LL. Self-organized periodic surface structures on ZnO induced by femtosecond laser. Applied Physics A: Materials Science and Processing 2010;100:517. [9] Jia TQ. Formation of nanogratings on the surface of a ZnSe crystal irradiated by femtosecond laser pulses. Physical Review B 2005;72:125429. [10] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Mechanisms of ultrafast laserinduced deep sub wave length gratings on graphite and diamond. Physical Review B 2009;79:125436. [11] Hsu EM, Crawford THR, Tiedje HF, Haugen HK. Periodic surface structures on gallium phosphate after irradiation with 150 fs 7 ns laser pulses at 800 nm. Applied Physics Letters 2007;91:111102. [12] Xia ZL, Shao JD, Fan ZX. Statistical approach to bulk inclusion initialized damage in films. Optics Communications 2006;265:620. [13] Wei CY, Shao JD, He HB, Yi K, Fan ZX. Mechanism initiated by nano absorber for UV nanosecond-pulse-driven damage of dielectric coatings. Optics Express 2008;16:3376. [14] Zhang DP, Wang CJ, Fan P. Influence of plasma treatment on laser-induced damage characters of HfO2 thin films at 355 nm. Optics Express 2009;17: 8246. [15] Zhang DP, Shao JD, Zhang DW. Employing oxygen-plasma post treatment to improve the laser-induced damage threshold of ZrO2 films prepared by the electron-beam evaporation method. Optics Letters 2004;29:2870. [16] Zhao YA, Wang T, Zhang DW, Shao JD, Fan ZX. laser damage accumulation effects of HfO2/SiO2 antireflective coatings. Applied Surface Science 2005;245:335. [17] Liang LP, Zhang L, Sheng YG. Studies on the laser-induced damage resistance of sol–gel derived ZrO2–TiO2 composite high refractive index films. Acta Physica Sinica 2007;56:3596. [18] Grosso D. Scandia optical coatings for application at 351 nm. Thin Solid Films 2000;368:116. [19] Xia ZL, Wang H, Xu Q. The stress relief mechanism in laser irradiating on porous films. Optical Communications 2012;285:70. [20] Wei CY, Shao JD, He HB, Yi K, Fan ZX. Mechanism initiated by nanoabsorber for UV nanosecond-pulse-driven damage of dielectric coatings. Optics Express 2008;16:3376. [21] Smith AV, Do BT. Bulk and surface laser damage of silica by picosecond and nanosecond pulses at 1064 nm. Applied Optics 2008;47:4812. [22] Du D, Liu X, Korn G, Squier J, Mourou G. Laser-induced breakdown by impact ionization in SiO2 with pulse widths from 7 ns to 150 fs. Applied Physics Letters 1994;64:3071. [23] Raether H. Surface plasmons on smooth and rough surfaces and on gratings. Springer-Verlag; 1988. [24] Huang M, Zhao FL, Cheng Y, Xu NS, Xu ZZ. Origin of laser-induced nearsubwavelength ripples: interference between surface plasmons and incident laser. ACS NANO 2009;3:4062.