Formation of Au Nanostructures through an Alumina Mask by Laser-AssistedDeposition

Size: px
Start display at page:

Download "Formation of Au Nanostructures through an Alumina Mask by Laser-AssistedDeposition"

Transcription

1 JOURNAL OF NANOSCIENCE AND NANOTECHNOLOGY Formation of Au Nanostructures through an Alumina Mask by Laser-AssistedDeposition Chongho Kim, 1 Jong Bae Park, 1 Hae-geun Jee, 1 Soon Bo Lee, 1 Jin-Hyo Boo, 1 Seong Kyu Kim, 1 Ji-Beom Yoo, 2 Jin Seung Lee, 3 and Haeseong Lee 4 1 Department of Chemistry, Sungkyunkwan University, Suwon , Korea 2 Department of Materials Engineering, Sungkyunkwan University, Suwon , Korea 3 Eonnam High School, Seoul , Korea 4 Korea Basic Science Institute, Jeonju Branch, Jeonju , Korea We report a new method to produce ordered arrays of metal nanostructures on substrates. The method employs a through-hole nanoporous alumina membrane as a mask that is attached onto the substrate, silicon in this study. The material of deposition, Au in this study, was provided by pulsed laser ablation of a target gold. At an earlystage of the deposition, a significant portion of Au penetrated the alumina through-holes and formed an ordered nanodot arrayon the silicon surface. At the later stage, the through-hole deposition was blocked bythe growth of Au film on the top surface of the alumina, so that the heights of the Au nanodots were limited to about 10 nm under current experimental conditions. Subsequent attempts to clean up the top surface of the alumina with a lower power laser illumination resulted in the formation of new nanostructures around the alumina pores, nanospheres, or nanorings, depending on the fluence of the laser and the duration of the cleanup. We will discuss the underlying mechanism of the formation of these nanostructures. Keywords: AAO, Au, Nanostructure, Laser Ablation, Pulsed Laser Deposition. 1. INTRODUCTION Anodized aluminum oxide (AAO) has been considered an attractive template for fabricating highly ordered nanostructures. 1 8 It can be easily produced from an aluminum sheet of high purity following several electrochemical steps: electropolishing; 1st anodization; etching; nd anodization. The AAO film formed on the aluminum substrate displays a two-dimensional array of hexagonal cells with cylindrical pores whose diameters and interpore distances are very uniform and adjustable in the range of a few tens to hundreds of nanometers. The grown pores are straight along the AAO surface normal. Thus, once the aluminum substrate and the barrier layer of AAO are removed, the AAO membrane can be used as a mask to grow materials ordered on the nanometer scale. In this study, we have tested the possibility of deposition under vacuum. The material to be deposited is provided by evaporation of a target and the through-hole AAO membrane acts as a mask for the ordered structure. This method would be advantageous over electrochemical deposition 11 when the deposited structure needs be physically separated from the AAO template. Such vacuum deposition through the AAO mask was first demonstrated Author to whom correspondence should be addressed. by the Masuda group 2b c using electron beam evaporation of Au or Ag. In their works, the size of the AAO mask was limited to 2 mm 2 mm in order to collect exclusively the evaporated metals flying normal to the AAO surface. The directionality of the impinging metals onto the AAO mask must be a crucial factor in this technique. Similar performance would be expected if the laser ablation method replaces e-beam evaporation. The advantage of laser ablation would be that it may produce evaporated materials with much higher kinetic energy, so that the substrate with the AAO mask can be located closer to the target. In this way a faster deposition onto a wider area would be possible. The disadvantage would be that more often than not large clusters of laser-ablated material may block the pores of the AAO mask. The purpose of this paper is to investigate the performance of laserablated deposition through the AAO mask. In this study we chose Au for the deposited material, since it has a very high absorption cross-section at the laser wavelength used (532 nm). While the laser-ablated deposition was carried out, the upper surface of the AAO mask was covered with the evaporated material at some early stage of the process, which eventually resulted in blocking the pores. One method to clean up the covered material would be using secondary laser pulses to induce evaporation. However, instead of the covered material being completely cleaned up, new J. Nanosci. Nanotech. 2004, Vol. 5, No by American Scientific Publishers /2004/01/001/007/$ doi: /jnn.2004.j 1

2 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition J. Nanosci. Nanotech. 2004, 5, 1 7 ordered nanostructures around the AAO pores were found to form, which were sensitive to the laser power or the duration of the cleanup. We will discuss this process, too. 2. EXPERIMENTS We first synthesized an AAO mask from a 0.1-mm-thick aluminum sheet of % purity, following a wellknown procedure of the two-step anodization Adjusting the DC applied voltage (40 V), the temperature (17 C) of the electrolyte solution (0.3 M oxalic acid), and the growth time during the anodization processes (30 min for the second step), we were able to routinely produce AAO with uniformly spaced pores with a diameter of about 30 nm and a depth of about 2 m. Then, a porewidening process (30 C, 0.1 M phosphoric acid, 30 min) was followed to increase the regularity of the pore pattern. The final pore diameter was 50 nm. Then the aluminum substrate was removed with a saturated HgCl 2 solution, and the AAO barrier layer was etched away by a1m NaOH solution to form the through-hole AAO membrane to be used as a mask. The laser-assisted deposition chamber is schematically drawn in Figure 1. The chamber was maintained at the BS B L M M Si AAO Aperture Fig. 1. Drawing of the laser-assisted deposition chamber, the view from the top: 1, the target holder; 2, the substrate holder; 3, the ablation laser path; 4, the clean-up laser path; 5, the ion gun; BS, a beam splitter (10% reflection); M, mirrors; L, 30 cm focal length lens; B, laser beam block. The figure inside the dotted box gives more details of the substrate holder. The AAO mask was 1 cm 1 cm. The diameter of the aperture in the pressing cover was 8 mm in this study. base pressure lower than 10 5 Torr. Inside the chamber, a pellet of gold (99.99%) was mounted onto a target holder for the laser ablation. The target holder revolved at the speed of 10 cycles/min to produce a uniform plume throughout the ablation process. The substrate holder was positioned face-to-face from the target holder. The distance between the two holders was adjustable by translating the substrate holder. Results of this report were obtained with the distance of 2 cm. A shutter (not shown in Fig. 1) was placed in front of the substrate holder in order to block the target contaminants at the initial stage of ablation. In the substrate holder, a substrate Si(100) covered with the through-hole AAO mask was mounted. The Si wafer, without removing the natural oxide layer, was cleaned by sonication. The AAO mask, cut 1 cm 1 cm, was placed on the Si substrate and was pressed by a 1-mm-thick metal cover. The pressing cover had an aperture of 0.8-cm diameter and was bolted onto the mount (the dotted box of Fig. 1). In the laser-ablated deposition process, the target gold was ablated by 7-ns, 532-nm pulses from a Q-switched Nd:YAG laser (Continuum, Surelite-II) operating at a 10 Hz repetition rate. The laser beam was focused onto the target by a lens of 30-cm focal length. The focal diameter was determined in a conventional way to approximately 100 m, judged from a microscope image of a single shot burn of the target. In this way, the fluence of the ablation laser of this study was determined to be 60 J/cm 2, unless otherwise specified. In the second or the clean-up step, much weaker laser pulses, obtained by splitting a fraction of the ablation pulses, were illuminated without focusing onto the surface of the Au-deposited AAO mask. The spatial distribution of the clean-up laser intensity throughout the exposed area (0.8-cm diameter) of the AAO mask is almost flat. We used the cleanup laser fluence between 50 and 100 mj/cm 2. Higher clean-up fluence caused the damage of the AAO surface. The morphologies of the AAO and the nanostructures of this study were investigated with a field emission SEM (JEOL, JSM890) and an AFM (Digital Instrument, Nanoscope IV). In the images of the top surface of the AAO, each cell has a curved shape as indicated in Figure 2. Each cell of the bottom surface of the AAO is also curved, but to a much smaller degree. Here we define the bottom of the AAO as the side where the alumina barrier layer was etched away. Within the scan range (20 m) of the AFM, the rms roughness of the altitudes on the top and the bottom surfaces were determined to be 18 and 5.5 nm, respectively. These values of the roughness reflect the average curvatures of the AAO cells and imply that the altitude variance due to the grain domains is negligible. We were not able to measure the surface roughness in the longer range with the SEM or the AFM. 2

3 J. Nanosci. Nanotech. 2004, 5, 1 7 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition (a) Au 1 Al 2 O 3 2 Si (b) Laser Illumination 4 SKKU SEI 15.0 kv nm WD 8.0 mm (a) 3 low power prolonged illumination 7.0 Section Analysis high power Fig. 2. Schematics of the laser-assisted deposition of this study. (a) the through-hole deposition and film formation process; (b) the cleanup process. 3. RESULTS AND DISCUSSIONS 3.1. Laser-Ablated Deposition through an AAO Mask In order for the laser-ablated particles to penetrate the through-hole AAO mask effectively, the flight velocity must be high and its direction must be normal to the substrate. These conditions are best met without background gas on the laser ablation. The high flight velocity of the particle is achieved by a high enough fluence of the ablation pulse. 12 For example, in the work of Irissou et al., 13 where 17-ns, 248-nm laser pulses were used to ablate an Au target, the fluence of 4 J/cm 2 induced about 9 km/s flight velocity when there was no background gas. On the other hand, using too high a laser fluence induced formation of large clusters in the plasma of too high density. 14 Thus, the laser fluence (60 J/cm 2 of our experiments must give high enough flight velocity, on the order of 10 1 km/s. At this laser fluence, the plume generated by the ablation of gold was visible up to about 1 cm. Accordingly, the substrate was positioned 2 cm away to avoid a high density of Au particles. Under this condition, the morphology of the deposited Au film was found uniform over the exposed area of the AAO surface with a roughness of less than 10 nm. Under the deposition condition mentioned above, we first looked into the structure of the substrate from the first step after a laser ablation of 5 min, structure 2 of Figure 2. When the Si substrate after the AAO mask was detached was investigated with the SEM, a twodimensional array of Au dots following the pattern of the AAO mask was clearly seen (Fig. 3a). The EDAX mode of the SEM demonstrated that the dots were undoubtedly Au. This array of Au dots was extended over the exposed µm nm (b) µm Fig. 3. Array of Au nanodots formed on the Si surface by laser-ablated deposition through the AAO mask. (a) SEM image; (b) AFM image. Note that tiny (5 10 nm in size) and random features in the SEM image are fragments of the Pt coating, which was used for a better image of the SEM. area, (0.4 cm 2 ). The average diameter of the dots was about 60 nm, slightly larger than the average pore diameter of the AAO mask. The sectional analysis of the Au dots, investigated with the AFM under the tapping mode (Fig. 3b), showed that the average height of the dots was about 5 nm with the standard deviation of 1 nm. The reason the diameter of the dots was slightly larger than that of AAO pores would be attributed to the imperfect contact of the silicon surface and the AAO bottom surface, which unblocked the deposited Au to disperse laterally on the silicon surface. The degree of the contact between the AAO mask and the silicon substrate is one of the critical aspects of this technique to produce a well-ordered array. The imperfect contact could be produced by the inherent long-range roughness of the AAO mask or the Si surface and by the small roughness due to the curvature of the AAO cells. Until now, we have not treated the roughness of the AAO surface, but we expect that mechanical polishing or ion milling the AAO surface would improve the contact. Throughout many repeated trials of the first-step deposition, we have not been able to deposit Au dots with heights greater than 10 nm. To investigate the possibility that the deposited Au was retained inside the AAO pores after the AAO was detached, we looked into the inner walls of AAO pores with the SEM. Small amounts of Au in the form of tiny Au clusters or in the form of a thin film remained on the walls of AAO pores. However, the AAO pore channels were unblocked except the top part. Therefore, the primary reason for the limit of the grown height is that the top parts of the AAO pores were clogged after 3

4 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition J. Nanosci. Nanotech. 2004, 5, 1 7 some time on the laser ablation. We investigated the SEM images of the AAO top surface following several ablation intervals. At an early stage of the deposition (1 min), tiny Au clusters with diameters of a few nanometers were found on the AAO top surface, similar to the observations by Hu et al. 14 They explained that the tiny clusters are formed while the initially hot and mobile particles are cooled to solidify on the substrate. As the ablation time increased, the amount of the Au clusters on the AAO surface increased and began to aggregate when their density was high enough. At 5 min of ablation, the AAO pores began to be clogged by the aggregated Au, and after that Au grew as a film on the AAO surface. After 10 min of laser ablation, the film thickness reached about 30 nm. Therefore, the formation of Au nanodots at the boundary of the silicon and the AAO bottom was mostly completed within the first 5 min of deposition Nanostructure Formation on AAO by the CleanupProcess In order to open the clogged pores and deposit more Au on silicon, we attempted to clean up the deposited Au on the AAO surface by evaporating with a lower power laser illumination. This method of cleanup would be plausible since Au has a much higher absorption cross-section than alumina. For the cleanup, the AAO top surface, covered with 30-nm-thick Au film formed by laser ablation for 10 min, was illuminated. However, instead of Au being evaporated, new ordered nanostructures around the pores developed (structures 4 and 5 of Fig. 2 or Fig. 4). Parts a and b of Figure 4 are the SEM images of the nanospheres formed after 5-min illumination by the 55 mj/cm 2 cleanup laser. The average diameter of the spheres was about 60 nm, close to that of the AAO pores. From a simple calculation of the volumes of the spheres and the initial 30-nm-thick film, we figure that at least two-thirds of Au remained on the AAO top surface to form the spheres. Figure 4c is the case when the 80 mj/cm 2 cleanup laser was illuminated for 5 min. In this case, while more Au disappeared into vacuum or into the AAO pores, the remaining Au formed rings around the AAO pores. A similar nanoring structure was also produced if the 55 mj/cm 2 cleanup laser was illuminated for 10 min or longer. In other words, the nanosphere could be an intermediate structure to form the nanoring. The formation of the nanospheres and the nanorings was investigated with SEM at several intervals of the 55 mj/cm 2 cleanup illumination (Fig. 5). The Au film structure disrupted after a short cleanup illumination. After 3 min, some Au spheres began to appear around several pores. The disrupted Au and the Au spheres coexisted, while a portion of the latter increased as the cleanup illumination continued. After 5 min, the Au spheres covered the whole surface and stayed stable for about a minute of follow-up illumination. As the illumination time was prolonged, the central part of some spheres vaporized in an explosive fashion. After 10 min, the nanoring structure dominated the AAO surface. While these sequential steps were observable separately in the case of the lower power illumination, we were not able to observe any homogeneous intermediate structures like nanosphere when 80 mj/cm 2 was used for the illumination. In this case, as the transformation between the structures must be faster, the various forms of nanostructures coexisted in mixtures. SKKU SEI 15.0 kv 100, nm WD 8.0 mm SKKU SEI 15.0 kv 100, nm WD 8.0 mm (a) SKKU SEI 15.0 kv 100, nm WD 8.2 mm (c) Fig. 4. SEM images of the Au nanostructures formed on AAO when the cleanup laser was illuminated. (a, b) The laser fluence was 55 mj/cm 2 ; (c) the laser fluence was 80 mj/cm 2. The illumination time was 5 min. Note that tiny (5 10 nm in size) and randomly shaped features in the SEM image are fragments of the Pt coating, which was used for a better image of the SEM. (b) 3.3. Possible Mechanism in the CleanupProcess The mechanism of forming the new nanostructures, induced by the cleanup laser, is not clearly understood but must be related to the thermodynamical effects. Similar droplets or wavelike structures on the target surface are known to form when the laser fluence is close to the ablation threshold Bennett et al. 16 have explained that the anisotropic difference in the thermal expansion between the melted phase and the solid phase may cause an inertia force to form such surface topography on the target. However, the periodicity of the waves and the size of the droplets observed in their works were on the order of micrometers in the case of Au, while our new structures are on the nanometer scale. We also point out that the fluences of the cleanup laser that induced these nanostructures are an order of magnitude lower than the ablation threshold of bulk Au. In order to know the degree of the thermal effects, we estimate the temperature jump profile ( T of the Au film 4

5 J. Nanosci. Nanotech. 2004, 5, 1 7 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition Acc.V Spot Magn Det WD 200 nm SEI 15.0 kv nm WD 8.0 mm 15.0 kv x SE SKKU SEI 15.0 kv nm WD 8.0 mm (a) (b) (c) Acc.V Spot Magn Det WD 15.0 kv x SE (d) 200 nm SKKU SEI 15.0 kv nm WD 7.9 mm (e) Acc.V Spot Magn Det WD 15.0 kv x SE Fig. 5. SEM images of the Au nanostructures formed on AAO when the cleanup laser of 55 mj/cm 2 was illuminated for durations as follows: (a) 3 min; (b) 4 min; (c) 5 min; (d) 7 min; (e) 8 min; (f) 10 min. induced by the cleanup laser, under a reasonable assumption that the lateral heat diffusion during the laser pulse is small compared with the laser beam diameter, as in eq. (1) t T = K C 2 1 I0 F t d (1) Here,, K,, and C are light absorptivity at 532 nm, thermal conductivity, density, and heat capacity, respectively. I 0 and F t are the fluence and the temporal shape function of the laser pulse, respectively. K,, and C values for Au are given in a handbook as 3.17 W cm 1 K 1, 19.3 g cm 3, and J g 1 K 1, respectively. For F t, a simple square function with 7-ns width is assumed. Then, for the 55 mj/cm 2 cleanup laser fluence, T = (210 C). The absorptivity of the gold film is most uncertain in this calculation. A simple bulk value of 0 4 estimated under the Drude formalism gives a temperature jump of 84 C, reaching well below the melting temperature of gold (1064 C). Even considering that the deposited Au film must have higher absorptivity because of the plasma resonance band near 500 nm and the rough morphology, the peak temperature induced by the cleanup laser of 55 mj/cm 2 would not exceed 200 C. There are many reports that the melting temperature of gold particles decreases rapidly from the bulk value as its diameter decreases below 50 nm Similarly, it is plausible that the melting threshold of a thin film is significantly lower than that of the bulk as the thermal diffusion is restricted. This consideration is supported by several findings of other laser ablation experiments. For example, in the works of Matthias et al., 21 the threshold fluence for 0 (f) 200 nm laser ablation (14 ns, 248 nm) of Au films decreased linearly from 1.5 J/cm 2 when the film thickness was 1.1 m to 0.2 J/cm 2 when the thickness was 0.1 m. The threshold fluence for the optical damage, which these authors assumed to be the melting threshold, decreased linearly from 0.4 J/cm 2 when the film thickness was 1.1 m to 0.1 J/cm 2 when the thickness was 0.1 m. The 1.1 m was considered to be the thermal diffusion length of Au. In a similar report by Rosenfeld et al., 22 using a 20-ns, 248-nm laser, the ablation threshold of Au film decreased from 2.0 J/cm 2 when the film thickness was 1.0 m to 0.25 J/cm 2 when the thickness was 0.05 m. The threshold fluences of these reports seem to be higher than our 55 mj/cm 2 that induced the formation of nanospheres. However, as we used laser pulses with shorter duration and as the threshold fluence for melting is different from the threshold fluence for ablation or optical damage, their numbers mean only indirect references, suggesting that the threshold fluence for melting of the 30-nm Au film is about 4 8 times lower than that of the bulk. This consideration along with a rough calculation with eq. (1) puts 55 mj/cm 2 close to the threshold fluence for melting of the 30-nm Au film. Thus, we believe that the formation of the nanostructures of Figures 4 and 5 was initiated by the melting of the Au film. In the beginning of the illumination, the film structure is disrupted by melting. Then, the mobilized Au may reconstruct its form until it is solidified by cooling. As the sequence of melting, reforming, and solidifying is repeated for every cleanup laser pulse, Au atoms gather around the pores due to higher adhesive interaction at the pore perimeters and form spheres in order to decrease the 5

6 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition J. Nanosci. Nanotech. 2004, 5, 1 7 surface tension. This transformation may occur by acceleration at a later stage as the melting threshold of the new form may become lower than that of the film structure as the new form is more confined. The nanospheres stay stable for a few minutes when the lower power laser is used but transform into the nanorings by evaporation at their centers for the prolonged illumination. From our experience with AAO, we believe the pore perimeters generally have stronger adhesive interaction with foreign atoms than the other parts of AAO. Therefore, it is not surprising that the Au atoms that were near the pore perimeters did not disappear. The formation of the nanorings from the nanospheres implies that the thermal energy induced by each laser shot was accumulated in the Au nanospheres. The fact that the thermal energy was not completely dissipated into the AAO for the 100-ms interval of laser pulses is somewhat surprising. This may be possible in our system, in which the contact area between the Au spheres and the AAO pores is so small, and shows an example of a peculiar phenomena in the nanometer domain. 4. SUMMARY AND CONCLUDING REMARK In this work, we showed that it is possible to deposit materials through a porous AAO mask using pulsed laser deposition in vacuum. Three forms of arrayed nanostructure are demonstrated: nanodots on a silicon surface, nanospheres, and nanorings around AAO pores. Each structure requires optimal experimental conditions. For the formation of nanodots on silicon, the fluence of the ablation laser should be to give maximum flight velocity but should be lower than the threshold of forming large clusters in the plume. The contact between the AAO mask and the silicon surface is also crucial. In our study, a metal cover plate with an aperture of a rather large diameter (0.8 cm) was used to press the AAO mask onto the silicon substrate in order to deposit as large an area as possible. For applications that need a smaller area of deposition, the size of the aperture may be reduced for better contact. The roughness of the AAO bottom surface can be another factor. Our AAO bottom surface was prepared after chemical etching and leaves room for better roughness, e.g., by ion milling. The nanodot array can find various applications. For example, we have succeeded in forming an array of a gold binding protein on the Au/Si substrate prepared by using the method of this study. 23 Cleaning up the covered Au from the AAO surface with lower power laser illumination was not successful. It instead developed new nanostructures around the AAO pores. Laser-induced melting followed by laternal flow and nucleation with strong adhesive interaction around the AAO perimeters is considered to have competed with vaporization to form such nanostructures. The formation of the Au nanospheres or the Au nanorings is possible since gold has a much higher absorption cross-section than AAO and since the thermal conductivity of AAO is very low. These structures would not form when a thin film of gold on AAO is simply heated. For the case of nanosphere formation, the initial film should be slightly thicker than that which gives the volume of spheres of the AAO pore diameter. The cleanup laser fluence should be slightly higher than that which induces melting of the corresponding thickness of the film. Prolonged illumination following nanosphere formation produces the nanoring structure, suggesting that thermal dissipation into the AAO pores is not efficient so that the accumulation of thermal energy is possible in the nanospheres. Acknowledgments: This work was supported by National R&D Project for Nano Science and Technology (Contract M B ) of MOST and SRC program (Center for Nanotubes and Nanostructured Composites) of KOSEF. References and Notes 1. (a) A.-P. Li, F. Müller, A. Birner, K. Nielsch, and U. Gösele, Adv. Mater. 11, 483 (1999). (b) M. Steinhart, J. H. Wendorff, A. Greiner, R. B. Wehrspohn, K. Nielsch, J. Schilling, J. Choi, and U. Gösele, Science 296, 1997 (2002). 2. (a) H. Masuda and K. Fukuda, Science 268, 1466 (1995). (b) H. Masuda and M. Satoh, Jpn. J. Appl. Phys. 35, L126 (1996). (c) H. Masuda, K. Yasui, and K. Nishio, Adv. Mater. 12, 1031 (2000). 3. (a) D. Routkevitch, A. A. Tager, J. Haruyama, D. Almawlawi, M. Moscovits, and J. M. Xu, IEEE Trans. Electron Devices 43, 1646 (1996). (b) J. Li, C. Padadopoulos, J. M. Xu, and M. Moscovits, Appl. Phys. Lett. 75, 367 (1999). 4. (a) C. R. Martin, Chem. Mater. 8, 1739 (1996). (b) J. Hulteen and C. R. Martin, J. Mater. Chem. 7, 1075 (1997). 5. Y. Kanamori, K. Hane, H. Sai, and H. Yugami, Appl. Phys. Lett. 78, 142 (2001). 6. D. Crouse, Y.-H. Lo, A. E. Miller, and M. Crouse, Appl. Phys. Lett. 76, 49 (2000). 7. (a) J. S. Suh, J. S. Lee, and H. Kim, Synth. Met. 123, 381 (2001). (b) J. S. Lee, G. H. Gu, H. Kim, J. S. Suh, I. Han, N. S. Lee, J. M. Kim, and G. S. Park, Synth. Met. 124, 307 (2001). 8. (a) M. J. Kim, T. Y. Lee, J. H. Choi, J. B. Park, J. S. Lee, S. K. Kim, J.-B. Yoo, and C.-Y. Park, Diamond Relat. Mater. 12, 870 (2003). (b) M. J. Kim, J. H. Choi, J. B. Park, S. K. Kim, J.-B. Yoo, and C.-Y. Park, Thin Solid Film 435, 312 (2003). (c) J. B. Park, Y. Kim, S. K. Kim, and H. Lee, Bull. Korean Chem. Soc. 25, 353 (2004). 9. H. Masuda, H. Yamada, M. Satoh, H. Asoh, M. Nakao, and T. Tamamura, Appl. Phys. Lett. 71, 2770 (1997). 10. S. Shingubara, O. Okino, Y. Sayama, H. Sakaue, and T. Takahagi, Jpn. J. Appl. Phys. 36, 7791 (1997). 11. (a) K. Nielsch, F. Müller, A. Li, and U. Gösele, Adv. Mater. 12, 582 (2000). (b) J. Choi, G. Sauer, K. Nielsch, R. B. Wehrsphon, and U. Gösele, Chem. Mater. 15, 776 (2003). 12. R. K. Singh and J. Narayan, Phys. Rev. B 41, 8843 (1990). 13. E. Irissou, B. L. Drogoff, M. Chaker, and D. Guay, Appl. Phys. Lett. 80, 1716 (2002). 14. C.-W. Hu, A. Kasuya, A. Wawro, N. Horiguchi, R. Czajika, Y. Nishina, Y. Saito, and H. Fujita, Mater. Sci. Eng. A217/218, 103 (1996). 6

7 J. Nanosci. Nanotech. 2004, 5, 1 7 Kim et al./formation of Au Nanostructures by Laser-Assisted Deposition 15. (a) R. Kelly and J. E. Rothenberg, Nucl. Instrum. Methods B 7/8, 755 (1985). (b) R. Kelly, J. J. Cuomo, P. A. Leary, J. E. Rothenberg, B. E. Braren, and C. F. Aliotta, Nucl. Instrum. Methods B 9, 329 (1985). 16. T. D. Bennett, C. Grigoropoulos, and D. J. Krajnovich, J. Appl. Phys. 77, 849 (1995). 17. J. F. Ready, Effects of High-Power Laser Radiation, Academic Press, New York (1971). 18. R. B. Hall, J. Phys. Chem. 91, 1007 (1987). 19. Ph. Buffact and J.-P. Borel, Phys. Rev. A 13, 2287 (1976). 20. T. Castro, R. Reifenberger, E. Choi, and R. P. Andres, Phys. Rev. B 42, 8548 (1990). 21. E. Matthias, M. Reichling, J. Siegel, O. W. Käding, S. Petzoldt, H. Skurk, P. Bizenberger, and E. Neske, Appl. Phys. A 58, 129 (1994). 22. A. Rosenfeld and E. E. B. Campbell, Appl. Surf. Sci , 439 (1996). 23. S. J. Lee, T. J. Park, J. P. Park, J. B. Park, T. Kim, S. K. Kim, Y. B. Shin, M.-G. Kim, B. H. Chung, H. Lee, S. K. Kim, and S. Y. Lee, submitted to Langmuir. 7