A p-polarized femtosecond laser beam was used to irradiate a UV grade fused silica to create microchannels, which are useful for producing optical gratings or micro fluidics devices. The laser irradiated surface was characterized using optical microscope, stylus profiler, SEM, XRD and TEM. A special technique was used to protect the laser irradiated surfaces in preparing cross-sectional TEM samples. The XRD spectra and TEM observation reveal that structure of the fused silica remain amorphous after the femtosecond laser irradiation.
© 2006 Optical Society of America
Laser treatment of surfaces has been used to induce crystallization of amorphous materials. For instance, excimer laser with pulse duration of about 20 ns is used to crystallize amorphous Si film for thin film transistor (TFT) applications [1–2]. The crystallization is attributed to melting of material and the spontaneous nucleation and crystal growth processes [1–3].
With the rapid advancement in laser technology, ultrashort pulse lasers such as femtosecond laser have become easily available nowadays. One of the major research areas is to study the beam-material interaction to understand how the ultrashort laser pulses affect surface morphology and chemistry . As the thermal diffusion length is proportional to the square root of the pulse duration, the femtosecond laser-induced thermal effect is expected to be negligible. However, in recent years femtosecond laser-induced crystallization has been reported in a wide range of materials, including Ge , Si [6,7], Ge2Sb2Te5 film  and Fe:LiNbO3 crystal . Furthermore, Kondo et al. [10,11] reported selective crystallization in a glass containing special ions such as Ag+ and Ce+ when irradiated with femtosecond laser (630 nm, 100 nJ, 500 shots).
Femtosecond laser can be used to make optical gratings in glass , but complete or partial crystallization of glass may affect its optical properties. Therefore, the project was carried out to understand whether or not femtosecond laser irradiation changes structure of a UV grade fused silica.
2. Experimental procedures
The material used was UV grade fused silica with thickness of 1 mm. It has thermal expansion coefficient of 0.55-0.57×10-6 °C-1, which makes it an ideal candidate for high repetition rate laser processing. The sample was irradiated with p-polarized femtosecond laser (Clark-MXR CPA-2010) in air. The laser emits 150 fs pulses of linearly polarized light at a central wavelength of 775 nm and with repetition rate of 1 kHz. Beam entrance diameter is 3.5 mm. The fused silica was placed at the focal plane of the focusing lens (focal length f=50 mm). The nominal laser beam spot diameter at the focal plane was about 14 µm. The laser beam was directed perpendicularly onto the sample surface with a fixed power of 31 mW. The single pulse energy was 31 µJ and laser fluence was calculated to be 20 J/cm2. Gratings were machined by moving the laser beam along the direction of polarization at a speed of 50 µm/s.
Morphology of the irradiated surface was examined using the optical microscope, scanning electron microscope (CamScan S360) and a stylus profiler (Taylor Hobson Precision Talyscan 150). Both non-irradiated and irradiated samples were characterized using EDX (energy dispersive X-ray microanalyzer), X-ray diffractometer (Philips Model PW1830) and TEM (JOEL 2010F transmission electron microscope). X-ray diffraction patterns were obtained using copper target as a source of X-ray with wavelength λ=1.5404 Å (Cu Kα1). The scanning angle was in the range of 10°–100° and scan speed of 0.02 °/s. The TEM samples were thinned to a thickness of 100 nm using PIPS (Gatan 691 Precision Ion Polishing System).
3. Results and discussion
Machining of microchannels was carried out using the femtosecond laser. After the laser irradiation, it was observed that the microchannels were straight with consistent cut width of 20 µm, as shown in Fig. 1. Characterization of the microchannels using a stylus profiler indicates that they are of V-shape with an average depth of about 5 µm, as shown in Fig. 2. No cracking was detected for the laser parameters used.
Energy dispersive spectra (EDS) obtained for non-irradiated and irradiated surface are compared in Fig. 3(a) and (b). As expected, there is no change in the chemical composition due to the femtosecond laser irradiation.
XRD was carried out to obtain X-ray diffraction spectra for the non-irradiated and femtosecond laser irradiated surfaces, as shown in Fig. 4(a) and (b). The absence of any sharp peaks in the broad band ranging from 0° angle to 100° angle indicates that the fused silica structure was amorphous and remained amorphous after the femtosecond laser irradiation.
However, it should be noted that the XRD results cannot completely exclude the possibility of occurrence of a small amount of crystals due to the laser irradiation. XRD technique may not detect crystalline diffractions if crystal size is small down to nanoscale or its amount is less than a certain critical level. For example, Revesz et al.  reported that there is not much difference in XRD patterns between Zr57Ti5-Cu20Al10Ni8 bulk metallic glass and the partially crystallized sample containing 15% of crystalline phases within the amorphous matrix.
To further examine whether femtosecond laser irradiation induced crystallization in the fused silica, great effort was made to prepare cross-sectional TEM samples from the laser irradiated surfaces. During preparation of the TEM samples, a special technique, as illustrated in Fig. 5, was used to protect the irradiated surfaces. Two cross-sectional slices of the microchannels with thickness 1 mm were first sectioned using diamond saw (Fig. 5(a)) and then glued face-to-face using M-bond adhesive (Fig. 5(b)). Afterwards, the sample was ground using progressively finer diamond abrasive paper to reduce its thickness to around 20 µm. Finally, the thin sample was transferred onto a TEM grid (Fig. 5(c)) for ion thinning to make it ready for TEM observation.
Examination of the cross-sectional samples was carefully carried out under the TEM to study effect of femtosecond laser irradiation on the structure of fused silica. Special attention was paid to the walls of the microchannels. The irradiated surfaces were all observed to be amorphous, as shown in Fig. 6(a) and (b).
Furthermore, electron diffraction was carried out and the diffraction patterns show that both non-irradiated and laser irradiated structures are amorphous, as shown in Fig. 7(a) and (b).
It should be noted that the study may represent the first direct TEM examination of femtosecond laser irradiated surface of fused silica using cross-sectional TEM samples.
Kasaai et al.  and Koubassov et al.  investigated surface modifications of fused silica samples and analyzed through X-ray diffraction and TEM the powder ejected during the femtosecond laser irradiation. Their results indicate the presence of melting and crystallization of fused silica.
However, it should be noted that they examined the powder rather than the femtosecond laser irradiated surface. Detection of crystallization in the powder ejected during femtosecond laser irradiation does not necessarily mean that crystallization must also occur on the laser irradiated surface. The results obtained by Kasaai et al.  can be used to support this argument. Kasaai et al.  observed crystallization in the ejected powder, but XRD spectrum of the femtosecond laser irradiated surface indicates that the fused silica remained amorphous.
Silica has strong glass forming ability, so it must be cooled very slowly from liquid state to produce crystalline phase. Rapid cooling favors noncrystalline formation since time is not allowed for long-range ordered arrangements of atoms. Femtosecond laser is known to produce very rapid heating and cooling rates, so it is not surprising to observe in the present study that the fused silica remains amorphous after femtosecond irradiation. However, femtosecond laser-induced crystallization and amorphization are complex physical processes which warrant further study in the future.
Microchannels that are free of microcracks were produced with femtosecond laser pulses. Extensive characterization of the non-irradiated and laser irradiated structures indicate that the femtosecond laser irradiation does not induce any crystallization in the UV grade fused silica.
Two of the authors (Zhou and Lim) acknowledge the financial support from A*STAR (Agency for Science, Technology and Research), Singapore, through the Strategic Research Program on “nanometrology for sustainable manufacturing growth”.
References and links
2. M. Hatano, S. Moon, M. Lee, K. Suzuki, and C.P. Grigoropoulos, “Excimer laser-induced temperature field in melting and resolidification of silicon thin films,” J. Appl. Phys. 87, 36–43 (2000). [CrossRef]
3. A.V. Demchuk and V.A. Labunov, “Surface morphology and structure modification of silicon layers indused by nanosecond laser irradiation,” Appl. Surf. Sci. 86, 353–358 (1995). [CrossRef]
4. H.X. Qian, W. Zhou, H. Y. Zheng, and G. C. Lim, “Morphological and chemical evolution on InP(100) surface irradiated with femtosecond laser,” Surf. Sci. 595, 49–55 (2005). [CrossRef]
5. A. Kapoor, A.P. Singh, and K.N. Tripathi, “Recrystallization of germanium surfaces by femtosecond laser pulse,” Opt. Laser Technol. 35, 87–97 (2003). [CrossRef]
6. J. Bonse, K.W. Brzezinka, and A.J. Meixner, “Modifying single-crystalline silicon by femtosecond laser pulses: an analysis by micro Raman spectroscopy, scanning laser microscopy and atomic force microscopy,” Appl. Surf. Sci. 221, 215–230 (2004). [CrossRef]
7. T.Y. Choi, D.J. Hwang, and C.P. Grigoropoulos, “Ultrafast laser-induced crystallization of amorphous silicon films,” Opt. Eng. 42, 3383–3388 (2003). [CrossRef]
8. G. Zhang, D. Gu, F. Gan, X. Jiang, and Q. Chen,“Femtosecond laser-induced crystallization in amorphous Ge2Sb2Te5 films,” Thin Solid Films. 474, 169–172 (2005). [CrossRef]
9. G.G. Cheng, J.D. White, L. Qing, Y.S. Wang, W. Zhao, and G.F. Chen, “Microstructure on surface of LiNbO3:Fe induced by a single ultra-short laser pulse,” Chin. Phys. Lett. 20, 1283 (2003). [CrossRef]
10. Y. Kondo, K. Miura, T. Suzuki, H. Intuye, T. Mitsuyu, and K. Hirao, “Three-dimensional arrays of crystallites within glass by using non-resonant femtosecond pulses,” J. Non-Cryst. Solids. 253, 143–156 (1999). [CrossRef]
11. Y. Kondo, T. Suzuki, H. Inouye, K. Miura, T. Mitsuyu, and K. Hirao, “Three-Dimensional Microscopic Crystallization in Photosensitive Glass by Femtosecond Laser Pulses at Nonresonant Wavelength,” Jpn. J. Appl. Phys. 37, L 94–L 96 (1998). [CrossRef]
12. H.Y. Zheng, W. Zhou, H.X. Qian, T.T. Tan, and G.C. Lim, “Polarisation-independence of femtosecond laser machining of fused silica,” Appl. Surf. Sci. 236, 114–119 (2004). [CrossRef]
13. A. Revesz, P. Ochin, P. Donnadieu, J.P. Simon, and P. Guyot, “Nanocrystallization in a Zr57Ti5Cu20Al10Ni8 bulk metallic glass,” Phil. Mag. Lett. 81, 767–775 (2001). [CrossRef]
14. M.R. Kasaai, S. Lagace, D. Boudreau, E. Forster, B. Muller, and S.L. Chin, “Creation of micro-holes on glass surface by femtosecond laser through the ejection of molten material,” J. Non-Cryst. Solids. 292, 202–209 (2001). [CrossRef]
15. V. Koubassov, J.F. Laprise, F. Theberge, E. Forster, R. Sauerbrey, B. Muller, U. Glatzel, and S.L. Chin, “Ultrafast laser-induced melting of glass,” Appl. Phys. A. 79, 499–505 (2004). [CrossRef]