Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Origin of periodicity in nanostructuring on thin film surfaces ablated with femtosecond laser pulses

Open Access Open Access

Abstract

We report physical processes responsible for the periodic nanostructure formation in femtosecond-laser ablation of thin film surfaces. It has been found that an initial random distribution of nanoscale ablation traces is periodically structured with an increase in superimposed laser pulses or fluence on diamond-like carbon film used as the target. The results show that the formation of periodicity can be attributed to the excitation of surface plasmon polaritons to induce the periodic enhancement of local fields in the surface layer. The estimated field period is in good agreement with the observed size of nanostructures.

©2008 Optical Society of America

1. Introduction

Intense ultrashort laser pulses have been demonstrated to be extremely effective for high energy-density excitation of solid surfaces and resulting precision processing of materials, because the ultrafast interaction can minimize undesirable thermal and mechanical effects on the target [1]. In such interaction of light with materials as the laser processing, the spatial resolution is usually limited to the order of laser wavelength by the diffraction limit. Despite this well-known limitation, the femtosecond (fs) laser ablation of a variety of materials has often been observed to form periodic nanostructures on the target surfaces [217], of which size is much smaller than that of so-called ripple structures [18,19]. The nanostructure formation observed has been stimulating a lot of studies to understand the physics of fs laser interaction with materials, since it is attractive for potential applications in nano-science and technology. The characteristic properties of nanostructuring have been studied as functions of laser parameters such as polarization, fluence, wavelength, and superimposed pulse number. The results obtained have shown that the nanostructure is usually produced with multiple laser pulses at low fluence around the ablation threshold, where the periodic structures have their wave vectors parallel to the laser E-field, and the structure size tends to be proportional to the laser wavelength λ used [4]. Based on the experimental results obtained, several authors have suggested possible physical mechanisms such as self-organization [3,7], a change in refractive index [5], the second harmonic generation [6,15], and Coulomb explosion [9]. However, none of them has been able to sufficiently illustrate the characteristic properties of nanostructure formation observed so far. It is clear that a more comprehensive and/or versatile physical model is required to understand the fascinating physics of nanostructuring.

The present authors and their collaborators have made a series of experimental studies to understand the interaction processes of nanostructuring on hard thin films such as diamond-like carbon (DLC) and TiN [4,8,12,2022]. In a recent experiment for DLC [21], the authors have found that the nanostructure formation is initiated on the swelled surface of which bonding structure is changed from DLC to glassy carbon (GC). The subsequent experiment using the patterned DLC target has clearly demonstrated that the ablation to create the nanostructure is preferentially induced with the help of a local field enhanced on the stripe with a high curvature [22]. The local field generation can account for the nanoscale ablation on the surface and its polarization dependence. For further understanding of nanostructuring, along with potential applications in nanoscience, one of the most important problems is the ultrafast interaction process to create the nanoscale periodicity on the film surface.

In this paper we report an experimental and theoretical study on the origin of the periodicity produced in the nanoscale ablation of DLC surface. It has been shown that the generation of periodicity can be attributed to the periodic enhancement of the local field through the excitation of surface plasmon polaritons (SPPs) in the surface layer. The observed period is shown to be in good agreement with the estimated plasmon wavelength, and the present picture reconciles with the observed properties of nanostructuring.

2. Experimental

In the experiment, we used the DLC film of 500 nm in thickness that was deposited on polished silicon substrates with a plasma-based ion implantation system. The surface roughness was measured to be less than 1 nm with a scanning probe microscope (SPM). For the ablation experiment, we used 800 nm, 100-fs pulses from a Ti:sapphire laser system operated at a repetition rate of 10 Hz. The linearly or circularly polarized laser beam was focused in air at normal incidence on the film surface with a spherical lens of 1000-mm focal length. The focal spot size is 190 µm in radius with the lowest-order Gaussian intensity distribution. The pulse energy used was in a range of 60–170 µJ with the corresponding fluence of F=50-150 mJ/cm2. At the low fluence used, a single fs laser pulse never induced visible ablation of the DLC. The superimposed number of laser pulses N on a target was 1–1000. The surface morphology was observed with a scanning electron microscope (SEM) and the SPM.

3. Results and discussion

 figure: Fig. 1.

Fig. 1. SEM images of DLC surfaces irradiated with the linearly polarized laser pulses of (a) N=10, (b) N=30, (c) N=100, (d) N=200, (e) N=500, and (f) N=1000 at a fixed fluence of F=100 mJ/cm2, where the laser polarization is horizontal.

Download Full Size | PDF

The morphological evolution in nanostructuring was observed as a function of N at a fixed fluence F. Figure 1 shows the SEM images of the surface ablated with the linearly polarized pulses of N=10-1000. For N=10, the visible ablation traces of which size is less than 10 nm start to be observed, whereas the surface image represents neither periodic structure nor polarization-dependent ablation trace. With increasing N up to 30, the randomly distributed ablation traces appear to be unified to form line-like traces with the spacing d or the period d=40-60 nm, where the width w of a trace is less than 10 nm with the length L=30-300 nm in the direction perpendicular to the laser polarization. For N=100-200, the line-like traces grow in depth and length with d ~120 nm and w ~40 nm. With a further increase in N to 500–1000, the broader and deeper ablation traces are structured to show the period d=140-170 nm. The characteristic evolution of nanostructuring was also observed with the circularly polarized laser pulses, where the initial random distribution of ablation traces was observed to form periodic granular traces of the diameter ϕ=40-100 nm with increasing N.

The result shown in Fig. 1 allows us to illustrate the possible interaction process of nanostructuring. In the most initial stage as N is increased, the bonding structure change from DLC to GC is randomly induced in the surface layer. This structural change certainly produces the surface roughness of nanometer size due to the random swelling of material. On the swelled surface with a high curvature, the local field is generated to enhance the incident E-field and initiate nanoscale ablation [22]. Once the ablation is locally induced, different high curvatures are created on the surface of nanometer size. Since the field enhancement should be larger in narrower spaces and valleys, subsequent laser pulses are able to extend the length of ablation traces to combine them. This picture would also be valid for the ablation with the circular polarization.

An increase in F is expected to provide almost the same effect on nanostructuring as increasing N [4,8,12,2022]. Figure 2 shows the typical SEM images of DLC surface ablated at F=80-150 mJ/cm2 of linearly and circularly polarized pulses. At the lowest fluence, the images shown in Figs. 2(a) and (d) do not represent any polarization dependence of ablation. At the higher fluences, the ablation traces are structured with d=60-120 nm for the linear polarization and ϕ=40-90 nm for the circular.

 figure: Fig. 2.

Fig. 2. SEM images of DLC surfaces irradiated with the linearly (upper three) and circularly (lower three) polarized laser pulses of N=100 at F=80 mJ/cm2 for (a) and (d), F=100 mJ/cm2 for (b) and (e), and F=150 mJ/cm2 for (c) and (f). The laser polarization for (a)–(c) is horizontal.

Download Full Size | PDF

Based on the results obtained, we attribute the origin of nanoscale periodicity to the excitation of SPPs that results from the coherent interaction of the incident laser field with free electrons created in the material. In the present experiment, the presence of surface roughness is crucial for the SPP excitation, because on a smooth surface the dispersion curve for the SPP lies below the photon line [23]. As mentioned before, the bonding structure change from DLC to GC produces a random distribution of nanometer-size roughness in the initial stage of ablation. This surface roughness much smaller than λ0 allows the incident E-field to coherently couple with the collective oscillations of free electrons to excite the SPPs in the surface layer. The SPP has a periodic change in the surface charge or resulting local field amplitude to periodically initiate the nanoscale ablation of the surface.

The plasmon wavelength λsp can be estimated by referring to the experimental results. First we evaluated the local field E local spontaneously generated on the curved DLC surface. For this we made the ablation experiment using the same patterned target as in our previous study [22]. Briefly, the target substrate made of Si was patterned with arrays of nanometer-size stripes, which were fabricated by electron-beam lithography and liftoff process. The patterned substrate was coated with 900-nm thick DLC film. A single stripe on the DLC surface was 500–600 nm in width and ~100 nm in height. For this DLC target, we observed the nanostructure formation as a function of F to quantitatively appraise the field enhancement on the stripe. Figures 3(a) and (b) show examples of the SPM image and its lateral scan for non-patterned flat and patterned stripe areas on a target, respectively. As seen with the scan traces, the nanostructure with a period d=100-120 nm is produced on the flat surface, while two grooves at a spacing d ~100 nm is created on the stripe ridge. It is noted that nanostructures of almost the same size d are produced at different fluences of F 1=100 mJ/cm2 for the flat area and F 2=70 mJ/cm2 for the stripe, whereas N=300 of fs laser pulses are used for both.

We may assume that the field intensity is the same on the two surfaces to produce nanostructures with the same period d. Then the observed lower value of F 2 or ΔF=F 1-F 2 suggests the generation of an additional local field on the stripe with a high curvature. The enhanced field intensity can be estimated from ΔF to be ΔE~4.0×106 V/cm along the laser E-field on the stripe surface. The additional field may be written as ΔE=E local cosθ, where E local is the local field spontaneously generated on the stripe surface with the angle θ between the laser E-field and the direction perpendicular to the stripe surface, as schematically illustrated in Fig. 3(c).

 figure: Fig. 3.

Fig. 3. Example of the SPM image and its lateral (x-z) scan for (a) the flat surface irradiated at the fluence F 1=100 mJ/cm2 and for (b) the patterned surface at F 2=70 mJ/cm2, where linearly polarized laser pulses of N=100 are used for both, and the arrows denote the polarization (x) direction, and (c) schematic diagram of the patterned DLC surface including the local field E local in the direction at the angle θ.

Download Full Size | PDF

The local field E local is generated by the surface charge coherently created and oscillated with the incident laser field. As shown in the previous study [22], we observed that ΔF was almost independent of F in the fluence region of interest. This suggests that the surface charge or the electron density N e produced for initiating the ablation is almost constant for a material, while an increase in F predominantly leads to an increase in the ablation depth and area. The electron density N e can be estimated from ΔE. Using the Gauss’s law [24], we have

ΔE=(ε1ε01)E0cos2θ,

where ε0 and ε1 are the dielectric constants of vacuum and the GC layer including the effect of free electrons, respectively. With the simple Drude model [25] in which the damping is ignored, ε 1 in the laser field is given by

ε1=ε0[εGC(ωp2ω2)],

where ε GC is the static relative dielectric constant of GC with ε GC=3.10+i3.17 [26], and ωp=[e 2 Ne/(ε0m)]1/2 is the plasma frequency with electron charge e and mass m. We approximate the measured cross section of the stripe, as shown in Fig. 3(c), by a semi-cylindrical shape with R ~400 nm in radius, 100 nm in height and 600 nm in foot-to-foot width. Then the ablation traces were found to develop at θ=78-86° on the stripe, as seen in our previous study [22]. With this angle, Eqs.(1) and (2) lead to N e=0.7 -6×1022 cm-3. The high electron density in excess of the critical value N e=1.7×1021 cm-3 at λ ~800 nm can be generated with the evanescent field rapidly decreasing in the surface layer.

The free electrons are predominantly produced in the GC layer to induce a large change in ε 1. For simplicity, we assume that the film surface before ablation consists of two layers of the upper GC and the lower DLC, as illustrated in the inset of Fig. 4. When the surface is weakly corrugated through the bonding structure change and subsequent ablation, the SPPs can be excited at the interfaces between the GC and DLC and also between air and the GC. The dispersion relation k sp=k 0[ε a ε b/(ε a+ε b)]1/2 has to be satisfied for the SPP excitation, where k sp and k 0 are the wave vectors of the surface plasmon and the incident light in vacuum [23], respectively, and ε a and ε b are the relative dielectric constants of two layers concerned. We calculated λ sp for the GC/DLC interface, using ε a=ε 1/ε 0 for the GC and ε b=ε 2=8.72 +i3.18 for the DLC [27]. Figure 4 shows the results of λ sp calculated as a function of N e in the GC layer, where λ sp for the air/GC interface is also shown for comparison. It is noted that the estimated electron density N e leads to λ sp=150-340 nm for the GC/DLC interface, whereas λ sp calculated for the air/GC interface is of the order of λ 0. The local ablation would be initiated at a period d ~ λ sp/2 with the help of the local field periodically enhanced by the SPPs. The period d=75-170 nm calculated for the GC/DLC interface is in good agreement with the observed size of nanostructure, where the smallest period d ~75 nm would be induced with ε a~-ε b at N e ~2×1022 cm-3.

 figure: Fig. 4.

Fig. 4. Plasmon wavelength calculated as a function of electron density for the GC/DLC interface (lower curve) and for the air/GC (upper curve).

Download Full Size | PDF

In spite of the simple surface structure assumed, the present model based on the excitation of SPPs is able to account for the observed characteristic dependences of nanostructuring on laser parameters. In the initial stage of ablation, the SPPs would be localized in the swelled surface. With increasing in N or F, the local field is periodically enhanced along the laser E-field direction to form the nanostructure through the ablation. The field enhancement should predominantly be developed on the surface area with higher curvature, as its period d ~ λ sp/2 is maintained. The enhanced-field period or resulting nanostructure size is proportional to the laser wavelength λ0 used, as given by the dispersion relation. A simulation study is now in progress to see the temporal evolution of local fields for nanostructuring.

4. Summary

We observed the initial stage of ablation for nanostructuring of the DLC film surface. The experimental results have shown that the origin of nanoscale periodicity can be attributed to the excitation of SPPs in the surface layer. The estimated period of local fields enhanced by the SPPs is in good agreement with the observed size of nanostructures, and the characteristic properties of nanostructuring observed so far reconcile with the present model.

Acknowledgments

The authors thank M.Obara, M.Ohtsu, and A.E.Kaplan for their helpful discussions and comments, and Y.Miyatani for his assistance in the experiment. This work was partially supported by the Grant-in-Aid for Scientific Research (A)18206010.

References and links

1. F. Korte, S. Adams, A. Egbert, C. Fallnich, A. Ostendorf, S. Nolte, M. Will, J. -P. Ruske, B. N. Chichkov, and A. Tünnermann, “Sub-diffraction limited structuring of solid targets with femtosecond laser pulses,” Opt. Express 7, 41–49 (2000). [CrossRef]   [PubMed]  

2. J. Bonse, H. Sturm, D. Schmidt, and W. Kautek, “Chemical, morphological and accumulation phenomena in ultrashort-pulse laser ablation of TiN in air,” Appl. Phys. A 71, 657–665 (2000). [CrossRef]  

3. F. Costache, M. Henyk, and J. Reif, “Modification of dielectric surfaces with ultra-short laser pulses,” Appl. Surf. Sci. 186, 352–357 (2002). [CrossRef]  

4. N. Yasumaru, K. Miyazaki, and J. Kiuchi, “Femtosecond-laser-induced nanostructure formed on hard thin films of TiN and DLC,” Appl. Phys. A 76, 983–985 (2003). [CrossRef]  

5. Q. Wu, Y. Ma, R. Fang, Y. Liao, Q. Yu, X. Chen, and K. Wang, “Femtosecond laser-induced periodic surface structure on diamond film,” Appl. Phys. Lett. 82, 1703–1705 (2003). [CrossRef]  

6. A. Borowiec and H. K. Haugen, “Subwavelength ripple formation on the surfaces of compound semiconductors irradiated with femtosecond laser pulses,” Appl. Phys. Lett. 82, 4462–4464 (2003). [CrossRef]  

7. F. Costache, M. Henyk, and J. Reif, “Surface patterning on insulators upon femtosecond laser ablation,” Appl. Surf. Sci. 208–209, 486–491 (2003). [CrossRef]  

8. N. Yasumaru, K. Miyazaki, and J. Kiuchi, “Glassy carbon layer formed in diamond-like carbon films with femtosecond laser pulses,” Appl. Phys. A 79, 425–427 (2004). [CrossRef]  

9. Y. Dong and P. Molian, “Coulomb explosion-induced formation of highly oriented nanoparticles on thin films of 3C-SiC by the femtosecond pulsed laser,” Appl. Phys. Lett. 84, 10–12 (2004). [CrossRef]  

10. P. Rudolf and W. Kautek, “Composition influence of non-oxidic ceramics on self-assembled nanostructures due to fs-laser irradiation,” Thin Solid Films 453–454, 537–541 (2004). [CrossRef]  

11. G. Daminelli, J. Krüger, and W. Kautek, “Femtosecond laser interaction with silicon under water confinement,” Thin Solid Films 467, 334–341 (2004). [CrossRef]  

12. N. Yasumaru, K. Miyazaki, and J. Kiuchi, “Fluence dependence of femtosecond-laser-induced nanostructure formed on TiN and CrN,” Appl. Phys. A 81, 933–937 (2005). [CrossRef]  

13. J. Bonse, M. Munz, and H. Sturm, “Structure formation on the surface of indium phosphide irradiated by femtosecond laser pulses,” J. Appl. Phys. 97, 013538 (2005). [CrossRef]  

14. X. C. Wang, G. C. Lim, F. L. Ng, W. Liu, and S. J. Chua, “Femtosecond pulsed laser-induced periodic surface structures on GaN/sapphire,” Appl. Surf. Sci. 252, 1492–1497 (2005). [CrossRef]  

15. T. Q. Jia, H. X. Chen, M. Huang, F. L. Zhao, J. R. Qiu, R. X. Li, Z. Z. Xu, X. K. He, J. Zhang, and H. Kuroda, “Formation of nanogratings on the surface of a ZnSe crystal irradiated by femtosecond laser pulses,” Phys. Rev. B 72, 125429 (2005). [CrossRef]  

16. R. Le Harzic, H. Schuck, D. Sauer, T. Anhut, I. Riemann, and K. König, “Sub-100 nm nanostructuring of silicon by ultrashort laser pulses,” Opt. Express 13, 6651–6656 (2005). [CrossRef]   [PubMed]  

17. T. Tomita, K. Kinoshita, S. Matsuo, and S. Hashimoto, “Distinct Fine and Coarse Ripples on 4H-SiC Single Crystal Induced by Femtosecond Laser Irradiation,” Jpn. J. Appl. Phys. 45, L444–L446 (2006). [CrossRef]  

18. Z. Guosheng, P. M. Fauchet, and A. E. Siegman, “Growth of spontaneous periodic surface structures on solids during laser illumination,” Phys. Rev. B 26, 5366–5381 (1982). [CrossRef]  

19. J. F. Young, J. S. Preston, H.M. van Driel, and J. E. Sipe, “Laser-induced periodic surface structures. II. Experiments on Ge, Si, Al, and brass,” Phys. Rev. B 27, 1155–1172 (1983). [CrossRef]  

20. K. Miyazaki, N. Maekawa, W. Kobayashi, N. Yasumaru, and J. Kiuchi, “Reflectivity in femtosecond-laser-induced structural changes of diamond-like carbon film,” Appl. Phys. A 80, 17–21 (2005). [CrossRef]  

21. G. Miyaji and K. Miyazaki, “Ultrafast dynamics of periodic nanostructure formation on diamondlike carbon films irradiated with femtosecond laser pulses,” Appl. Phys. Lett. 89, 191902 (2006). [CrossRef]  

22. G. Miyaji and K. Miyazaki, “Nanoscale ablation on patterned diamondlike carbon film with femtosecond laser pulses,” Appl. Phys. Lett. 91, 123102 (2007). [CrossRef]  

23. H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings (Springer-Verlag, Berlin, 1988), p.5.

24. J. D. Jackson, Classical Electrodynamics (3rd ed., John Wiley & Sons, New York,1999), Sec.1.3.

25. C. V. Shank, R. Yen, and C. Hirlimann, “Time-Resolved Reflectivity Measurements of Femtosecond-Optical-Pulse-Induced Phase Transitions in Silicon,” Phys. Rev. Lett. 50, 454–457 (1983). [CrossRef]  

26. M. W. Williams and E. T. Arakawa, “Optical properties of glassy carbon from 0 to 82 eV,” J. Appl. Phys. 43, 3460–3463 (1972). [CrossRef]  

27. S. A. Alterovitz, N. Savvides, F. W. Smith, and J. A. Woollam, Handbook of Optical Constants of Solids, edited by E. D. Palik and G. Ghosh (Academic, London,1997), p.837–852.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (4)

Fig. 1.
Fig. 1. SEM images of DLC surfaces irradiated with the linearly polarized laser pulses of (a) N=10, (b) N=30, (c) N=100, (d) N=200, (e) N=500, and (f) N=1000 at a fixed fluence of F=100 mJ/cm2, where the laser polarization is horizontal.
Fig. 2.
Fig. 2. SEM images of DLC surfaces irradiated with the linearly (upper three) and circularly (lower three) polarized laser pulses of N=100 at F=80 mJ/cm2 for (a) and (d), F=100 mJ/cm2 for (b) and (e), and F=150 mJ/cm2 for (c) and (f). The laser polarization for (a)–(c) is horizontal.
Fig. 3.
Fig. 3. Example of the SPM image and its lateral (x-z) scan for (a) the flat surface irradiated at the fluence F 1=100 mJ/cm2 and for (b) the patterned surface at F 2=70 mJ/cm2, where linearly polarized laser pulses of N=100 are used for both, and the arrows denote the polarization (x) direction, and (c) schematic diagram of the patterned DLC surface including the local field E local in the direction at the angle θ.
Fig. 4.
Fig. 4. Plasmon wavelength calculated as a function of electron density for the GC/DLC interface (lower curve) and for the air/GC (upper curve).

Equations (2)

Equations on this page are rendered with MathJax. Learn more.

Δ E = ( ε 1 ε 0 1 ) E 0 cos 2 θ ,
ε 1 = ε 0 [ ε G C ( ω p 2 ω 2 ) ] ,
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.