Tunable Metallization by Assembly of Metal Nanoparticles in Polymer

Publication Date (Web): August 27, 2005 ... Besides the creation of binary patterns, it is also possible to create grayscale patterns where the densit...
0 downloads 0 Views 681KB Size
Download PDF
9352

Langmuir 2005, 21, 9352-9358

Tunable Metallization by Assembly of Metal Nanoparticles in Polymer Thin Films by Photo- or Electron Beam Lithography Dehui Yin and Shin Horiuchi* Nanotechnology Research Institute, National Institute of Advanced Industrial Science and Technology, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan

Masamichi Morita and Atsushi Takahara Institute for Materials Chemistry and Engineering, Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Received April 28, 2005. In Final Form: July 24, 2005 The technique of patterning of surfaces with metal-rich structures on micro- or nanoscales was developed by assembling metal nanoparticles into a thin film of polymer in a controllable way. Palladium (Pd) nanoparticles were incorporated into a thin film of poly(methyl methacrylate) (PMMA) using palladium (II) bis(acetylacetonato), Pd(acac)2, as a precursor vaporized in a nitrogen atmosphere. Depending upon its dose, the irradiation of a PMMA film by UV light or an electron beam (EB) enhances its reducing capability against Pd(acac)2. This dependency on dose can be used to control the formation and assembly of Pd nanoparticles. Using this technique, binary patterns consisting of metal-rich and metal-poor regions in the polymer film can be created simply by irradiating the surface of the polymer through a binary photomask. Besides the creation of binary patterns, it is also possible to create grayscale patterns where the density of Pd nanoparticles can be tuned to provide shades of gray by the use of light with continuously modulated intensity. Because the electron beam also enhances the reducing power of PMMA against Pd(acac)2, it is thus possible to obtain highly metallized films with nanoscale pattern features. The PMMA film can be selectively removed by oxygen plasma treatment or by pyrolysis. Thus, highly metallized surfaces with binary or grayscale patterns can be obtained by selective removal of the PMMA films. The metallized regions possess relatively high resistivity against CF4 plasma compared to the bare silicon surface; therefore, the metallized surface patterns can be transferred onto the underlying silicon substrate by CF4 plasma treatment. Because of the nanosize effect of metal nanoparticles, the thermal treatment at 900 °C, which is significantly lower than the melting temperature of the bulk Pd, yields continuous metallic features by binding the assembled nanoparticles.

* Corresponding author. E-mail: [email protected]. Tel: 81-29-861-6281. Fax: 81-29-861-4773.

assemble them to create metallized structures with wellcontrolled spatial arrangements. One approach is the use of an organometallic polymer thin film as a resist for photoor EB lithography, which allows the creation of conductive and magnetic patterns with small feature sizes.14-17 Moreover, many other approaches have been employed to realize patterned metallized surfaces by introducing selfassembled monolayers (SAMs),18-21 nanoimprint lithography,22,23 soft lithography,24-28 nanosphere lithography,29-32 and so forth.

(1) Rotello, V. Nanoparticles: Building Blocks for Nanotechnology; Kluwer Academic/Plenum Publishers: New York, 2004; Chapter 5. (2) Grunes, J.; Zhu, J.; Anderson, E. A.; Somorjai, G. A. J. Phys. Chem. B 2002, 106, 11463. (3) Chen, M.-S.; Dulcey, C. S.; Brandow, S. L.; Leonard, D. N.; Dressick, W. J.; Calvert, J. M.; Sims, C. W. J. Electrochem. Soc. 2000, 147, 2607. (4) Dressick, W. J.; Chen, M.-S.; Brandow, S. L. J. Am. Chem. Soc. 2000, 122, 982. (5) Chen, M. S.; Brandow, S. L.; Dressick, W. J. Thin Solid Films 2000, 379, 203. (6) Huang, S.; Dai, L.; Mau, A. W. H. Adv. Mater. 2002, 14, 1140. (7) Xu, Y.; Sun, H.-B.; Ye, J.-Y.; Matsuo, S.; Misawa, H. J. Opt. Soc. Am. B 2001, 18, 1084. (8) Haes, A. J.; Zou, S.; Schatz, G. C.; Van Duyne, R. P. J. Phys. Chem. B 2004, 108, 6961. (9) Rotello, V. Nanoparticles: Building Blocks for Nanotechnology; Kluwer Academic/Plenum Publishers: New York, 2004; Chapter 6. (10) Delamarche, E.; Geissler, M.; Vichiconti, J.; Graham, W. S.; Andry, P. A.; Flake, J. C.; Fryer, P. M.; Nunes, R. W.; Michel, B.; O’Sullivan, E. J.; Schmid, H.; Wolf, H.; Wisnieff, R. L. Langmuir 2003, 19, 5923. (11) Delamarche, E.; Vichiconti, J.; Hall, S. A.; Geissler, M.; Graham, W.; Michel, B.; Nunes, R. Langmuir 2003, 19, 6567.

(12) Thurn-Albrecht, T.; Schotter, J.; Ka¨stle, G. A.; Emley, N.; Shibauchi, T.; Krusin-Elbaum, L.; Guarini, K.; Black, C. T.; Tuominen, M. T.; Russell, T. P. Science 2000, 290, 2126. (13) Chou, S. Y.; Wei, M. S.; Krauss, P. R.; Fischer, P. B. J. Appl. Phys. 1994, 76, 6673. (14) Cheng, A. Y.; Clendenning, S. B.; Yang, G.; Lu, Z.-H.; Yip, C. M.; Manners, I. Chem. Commun. 2004, 780. (15) Clendenning, S. B.; Aouba, S.; Rayat, M. S.; Grozea, D.; Sorge, J. B.; Brodersen, P. M.; Sodhi, R. N. S.; Lu, Z.-H.; Yip, C. M.; Freeman, M. R.; Ruda, H. E.; Manners, I. Adv. Mater. 2004, 16, 215. (16) Chan, W. Y.; Clendenning, S. B.; Berenbaum, A.; Lough, A. J.; Aouba, S.; Ruda, H. E.; Manners, I. J. Am. Chem. Soc. 2005, 127, 1765. (17) Cyr, P. W.; Rider, D. A.; Kulbaba, K.; Manners, I. Macromolecules 2004, 37, 3959. (18) Zangmeister, C. D.; van Zee, R. D. Langmuir 2003, 19, 8065. (19) Geissler, M.; Wolf, H.; Stutz, R.; Delamarche, E.; Grummt, U.W.; Michel, B.; Bietsch, A. Langmuir 2003, 19, 6301. (20) Mendes, P. M.; Jacke, S.; Critchley, K.; Plaza, J.; Chen, Y.; Nikitin, K.; Palmer, R. E.; Preece, J. A.; Evans, S. D.; Fitzmaurice, D. Langmuir 2004, 20, 3766. (21) Carvalho, A.; Geissler, M.; Schmid, H.; Michel, B.; Delamarche, E. Langmuir 2002, 18, 2406.

Introduction The patterning of surfaces with metal-rich structures is of special interest because it offers the possibility of fabricating functional devices with useful catalytic,1-6 optical,7 sensing,8,9 electrical,10,11 and magnetic properties.12,13 To fulfill their expectations, there is great motivation to use nanoparticles as building blocks and

10.1021/la0511485 CCC: $30.25 © 2005 American Chemical Society Published on Web 08/27/2005

Metallization of Thin Films by UV or EB Lithography

We have developed a simple dry process for the synthesis and assembly of metal nanoparticles in polymer films through the reduction of a metal complex used as a precursor.33-37 Palladium (II) bis(acetylacetonato), Pd(acac)2, is vaporized in a nitrogen atmosphere at 180 °C and then absorbed into a polymer film, where the metal complex is reduced to form Pd nanoparticles. The Pd nanoparticles can be assembled in block copolymer nanodomains that produce 2D37 and 3D34,35 arrangements of metal nanoparticles within the films. This is caused by the differences in the reduction capabilities among the components of the block copolymers. The metal complex is selectively reduced in the phase of the component that has stronger reduction power in a diblock copolymer film; as a consequence, the metal nanoparticles are selfassembled to make a periodic formation. In addition to this nanoscale arrangement of the metal nanoparticles by the bottom-up approach, we found that the irradiation of poly(methyl methacrylate) (PMMA) with UV light enhances the reduction power of the metal complex36 and thus provides a top-down approach for arranging the metal nanoparticles. Here the nanoparticles can be assembled in a pattern defined by a photomask. This top-down approach provides us with a much simpler way to assemble the metal particles in a lateral dimension although the size of the attainable pattern features is limited by the wavelength of the irradiating light. In this article, we report on our continued study of this simple method for creating assembled patterns of metal nanoparticles in thin PMMA films, which improves the pattern quality and reduces the patterning feature sizes. We also report on the fabrication of highly metallized surface structures by the selective removal of the polymer by introducing subsequent reactive ion etching (RIE) and thermal treatments. Experimental Section Materials. Pd(acac)2 was purchased from Johnson Matthey Materials Technology and was recrystallized from acetone before use. PMMA (Mn ) 350 000) and polystyrene (PS, Mn ) 200 000) were purchased from Aldrich Chemical Co. Reprecipitations of the polymers were done twice from methylene chloride solutions to methanol for purification. A PMMA film was prepared by spin casting on a Si wafer and on a cleaved NaCl single-crystal surface from toluene solution. The thickness of the films was measured by spectroscopic ellipsometer (M-220, Jasco Co., Japan). (22) Martensson, T.; Carlberg, P.; Borgstrom, M.; Montelius, L.; Seifert, W.; Samuelson, L. Nano Lett. 2004, 4, 699. (23) Yan, X.-M.; Contreras, A. M.; Bokor, J.; Somorjai, G. A. Nano Lett. 2005, 5, 745. (24) Cherniavskaya, O.; Adzic, A.; Knutson, C.; Gross, B. J.; Zang, L.; Liu, R.; Adams, D. M. Langmuir 2002, 18, 7029. (25) Guo, Q.; Teng, X.; Rahman, S.; Yang, H. J. Am. Chem. Soc. 2003, 125, 630. (26) Yan, Y.; Chan-Park, M. B.; Gao, J.; Yue, C. Y. Langmuir 2004, 20, 1031. (27) Yu, M.; Lin, J.; Wang, Z.; Fu, J.; Wang, S.; Zhang, H. J.; Han, Y. C. Chem. Mater. 2002, 14, 2224. (28) Ng, H. T.; Foo, M. L.; Fang, A.; Li, J.; Xu, G.; Jaenicke, S.; Chan, L.; Li, S. F. Y. Langmuir 2002, 18, 1. (29) Ormonde, A. D.; Hicks, E. C. M.; Castillo, J.; Van Duyne, R. P. Langmuir 2004, 20, 6927. (30) Kuo, C.-W.; Shiu, J.-Y.; Chen, P.; Somorjai, G. A. J. Phys. Chem. B 2003, 107, 9950. (31) Bullen, H. A.; Garrett, S. J. Nano Lett. 2002, 2, 739. (32) Haynes, C. L.; McFarland, A. D.; Smith, M. T.; Hulteen, J. C.; Van Duyne, R. P. J. Phys. Chem. B 2002, 106, 1898. (33) Nakao, Y. Chem. Lett. 2000, 766. (34) Horiuchi, S.; Sarwar, M. I.; Nakao, Y. Adv. Mater. 2000, 12, 1507. (35) Horiuchi, S.; Fujita, T.; Hayakawa, T.; Nakao, Y. Langmuir 2003, 19, 2963. (36) Horiuchi, S.; Fujita, T.; Hayakawa, T.; Nakao, Y. Adv. Mater. 2003, 15, 1449. (37) Hirooka, H. IBM J. Res. Dev. 1977, 3, 121.

Langmuir, Vol. 21, No. 20, 2005 9353 UV Lithography. UV irradiation on PMMA films was carried out using a spot light source, Lightningcure LC6 (Hamamatsu Photonics K. K.), attached with a flexible light guide emitting light from a mercury-xenon lamp having line spectra ranging from 240 to 400 nm. The irradiation was done by inserting optical filters that cut the light with a wavelength longer than 300 nm between the lamp and the light guide. EB Lithography. EB lithography was performed with a Hitachi S-4100 field-emission scanning electron microscope equipped with a beam-drawing stage (Beam Draw, Tokyo Technology Inc.). A 20 keV electron beam with a beam current of 50 pA was used to expose patterns designed by a computeraided design (CAD) system. Incorporation of Pd Nanoparticles into Films. The process developed by our group33-37 was employed to incorporate Pd nanoparticles into PMMA films. The details of this method were described in our previous papers. In this process, 10 mg of Pd(acac)2 and a polymer film were loaded into a glass vessel and were heated to 180 °C in vacuo in a nitrogen atmosphere. Pd(acac)2 was sublimated and absorbed into the polymer film to be reduced to the metallic Pd nanoparticles. Selective Removal of Polymers and Fabrication of Silicon Substrates. For the selective removal of PMMA from the films with the incorporated Pd nanoparticles, the films were treated with oxygen plasma using an in-house-fabricated tubular reactor that generated capacitively coupled plasma excited by a 13.56 MHz rf power source. The oxygen plasma was generated at a power density of 90 W and a gas pressure of 67 Pa. Pyrolysis under an argon gas flow at 550 °C for 30 min was also employed to remove the PMMA selectively from the films using a tube furnace. To fabricate a Si wafer, CF4 plasma treatment was employed at a power density of 10 W/cm2 and a gas pressure of 15 Pa using SAMCO compact etcher FA-1. Characterization. Transmission electron microscopy (TEM) micrographs were obtained using a LEO922 energy-filtering transmission electron microscope (LEO Eletronenmikroskopie GmbH, Germany) at an accelerating voltage of 200 kV, which integrates an Omega-type electron spectrometer. Statistical image analysis of the TEM micrographs was performed using digital image analysis software (AnalySIS, Soft Imaging System Co. Ltd., Germany). Scanning probe microscopy (SPM) images were taken with an SPA 300HV (Seiko Instruments Inc., Japan) operated in tapping mode. Scanning electron microscopy (SEM) images were taken with a Carl Zeiss ULTRA 55 at an accelerating voltage of 10 kV. X-ray photoelectron spectroscopy (XPS) spectra were acquired on a PHI Quantum spectrometer (ULVAC-PHI Inc., Japan) equipped with a hemispherical capacitor analyzer using monochromated X-rays from Al KR. The Pd nanoparticle content in the films was estimated from the residue in the thermogravimetric analysis (TGA) using a Perkin-Elmer Pyris 1 TGA at a heating rate of 20 °C/min under nitrogen flow.

Results and Discussion Micropatterning of Pd Nanoparticles in Thin PMMA Films via UV Photolithography. As reported in our previous papers,34,35 PMMA exhibits a unique behavior regarding the absorption and the reduction of the Pd(acac)2 vapor. PMMA absorbs the Pd(acac)2 vapor as the other polymers do, but it retards the reduction and the formation of the Pd nanoparticles. We speculate that the carbonyl group in the PMMA side chain stabilizes the Pd2+ ion by coordination, which retards its conversion to the Pd metallic state. It has been found that the irradiation of PMMA by UV light causes the elimination of this unique behavior, thus allowing the normal process of reduction as other polymers do. The chemistry that occurs in PMMA as the result of UV irradiation has been well studied because it has been developed as a high-resolution photoresist material.38 UV irradiation causes the scission of the main and side chains of PMMA and then yields volatile low-molecular-weight products. We attribute the enhancement of the reduction power of PMMA to the loss (38) Yin, D.; Horiuchi, S.; Masuoka, T. Chem. Mater. 2005, 17, 463.

9354

Langmuir, Vol. 21, No. 20, 2005

Figure 1. TEM images of micropatterned Pd nanoparticles in thin PMMA films prepared by UV lithography. (a) Lowmagnification image presenting the overall pattern in the film with 50 nm thickness. (b-d) High-magnification images showing the distribution of Pd nanoparticles in the irradiated areas (upper column) and the masked areas (lower column) of the PMMA films exposed to Pd(acac)2 vapor for (b) 15, (c) 30, and (d) 60 min.

of the ability to stabilize the Pd2+ ions due to the loss of the carbonyl groups of PMMA. Therefore, it is possible to assemble the Pd nanoparticles by conventional UV lithography in PMMA films. However, one problem with this approach is that some Pd nanoparticles are inevitably formed even in the masked regions, degrading the contrast between the exposed and masked regions. Ideally the region on the film corresponding to the opaque areas on the mask should result in metal-free rather than metalpoor regions, but such a clear-cut distinction is not feasible. The factors that influence the quality of the patterns are film thickness, irradiation dose of UV light, and the exposure time of Pd(acac)2 vapor. To obtain the conditions to produce the optimized assembled pattern, we first investigated the effects of these three factors on the pattern quality. Thin PMMA films with thicknesses of 30, 50, and 100 nm were prepared on cleaved NaCl single-crystal surfaces. After being irradiated with UV light at a dose of 3, 5, or 7 J/cm2 through a metal foil with square openings in direct contact with them, they were exposed to Pd(acac)2 vapor for times that varied from 15 to 90 min at 180 °C. The films were lifted off the substrates by dipping into water and were then mounted onto 600-mesh TEM grids for TEM observation. Figure 1a shows a low-magnification image presenting the overall pattern formed in the film with 50 nm thickness by exposure to Pd(acac)2 vapor for 60 min. As can be seen, the Pd nanoparticles with an average diameter of about 4 nm were assembled to replicate the pattern of the photomask, where the bright region corresponds to the masked area and the dark region corresponds to the irradiated area formed by the assembly of the Pd nanoparticles. Figure 1b-d shows the high-

Yin et al.

magnification images showing the dispersion of Pd nanoparticles in the irradiated areas (upper column) and in the masked areas (lower column). The exposure times to the vapor were 15 min for Figure 1b, 30 min for Figure 1c, and 60 min for Figure 1d. These micrographs indicate that the exposure time of the Pd(acac)2 vapor has a large effect on the dispersion of the Pd nanoparticles in the two areas. It is noticeable that 15 min was not enough to produce a clear pattern because the Pd particles in the irradiated area were assembled loosely and the number of particles was only a little larger than that in the masked area. With exposure times longer than 30 min, the assemblies of Pd particles in the two areas became significantly different; the irradiated areas were nearly covered with the Pd particles, whereas only a small number of Pd particles were dispersed in the masked area. Although the site selectivity of the Pd nanoparticles is not perfect, the difference between the irradiated and masked areas in terms of the degree of the Pd nanoparticles’ dispersion could be optimized by adjusting the film thickness, irradiation dose of UV light, and exposure time of Pd(acac)2 vapor. The influences of those three parameters on the Pd nanoparticle content in the two areas were evaluated by calculating the area fractions occupied by the nanoparticles in the TEM micrographs. Figure 2a shows the effect of the film thickness on the area fraction when the exposure time was 60 min and the UV irradiation dose was 7 J/cm2. This Figure indicates that a large difference in the fraction between the irradiated and the masked region can be attained when the PMMA film is thicker than 50 nm. Figure 2b shows the effect of the dose of UV irradiation on the film with a thickness of 50 nm when the exposure time was 60 min. With increasing dose of UV irradiation, the Pd nanoparticle content in the irradiated regions increases nearly linearly, and a significant difference between the two regions can be achieved with a dose of 7 J/cm2. Figure 2c shows the effect of the time of exposure of the Pd(acac)2 vapor to the film with 50 nm thickness when the dose of UV irradiation is 5 J/m2. This Figure indicates that the content in the masked region is maintained in the lower level within 60 min; however, the longer exposure time leads to an increase in the content in the masked region, resulting in a decrease in the patterning quality. We have evaluated the absorption of Pd(acac)2 and the formation of Pd nanoparticles in the PMMA films.34,35 It was found that the absorption of Pd(acac)2 vapor by PMMA increases with time as for the other polymers but the reduction and the formation of the Pd nanoparticles are retarded with a certain induction period. This induction period is rather critical and is around 60 min. Therefore, exposure times longer than the induction period will lead to the formation of many more particles even in the masked region of the

Figure 2. Area fraction of Pd nanoparticles in the irradiated (b) and masked areas (9) calculated from the TEM micrographs. (a) Effect of the PMMA film thickness when the dose of UV irradiation is 7 J/cm2 and the exposure time of Pd(acac)2 vapor is 60 min. (b) Effect of the dose of UV irradiation when the film thickness is 50 nm and exposure time of Pd(acac)2 vapor is 60 min. (c) Effect of exposure time of Pd(acac)2 vapor when the film thickness is 50 nm and the dose is 5 J/cm2.

Metallization of Thin Films by UV or EB Lithography

Figure 3. TEM micrograph showing the distribution of Pd nanoparticles in a thin microtomed section with a thickness of 50 nm from a free-standing PS film exposed to Pd(acac)2 vapor for 60 min.

PMMA film. Considering these results, we conclude that for the optimization of the patterning quality the film thickness should be in the range of 50 to 100 nm, the dose should be higher than 7 J/cm2, and the exposure time should be around 60 min. It is difficult to determine Pd nanoparticle content in the thin films. For the estimation of its content qualitatively, we compare the TEM images shown in Figure 1 with a TEM image taken from a free-standing film with a known Pd nanoparticle content. The Pd nanoparticle content can be estimated from the residue in the TGA measurement of a free-standing film if the matrix polymer is volatized completely. Figure 3 shows a TEM micrograph of a thin section of a PS free-standing film exposed to Pd(acac)2 vapor for 60 min, of which the content was estimated to be 5 wt %. It was confirmed that the pure PS film gave no residue. The thin section from the freestanding film was adjusted to 50 nm thickness by ultramicrotomy to compare with the images of the thin films under the same condition. The number-average diameter and the fraction of the area occupied by the particles of the image shown in Figure 3 are about 5 nm and 4.3%, respectively. By comparison with the thin spincoated films shown in Figure 1, it is apparent that the Pd nanoparticle contents produced in the UV-irradiated regions in the thin films are remarkably high (upper column in Figure 1c and d). In such thin films, the UVirradiated areas can be highly metallized because the

Langmuir, Vol. 21, No. 20, 2005 9355

metal complex vapor can be highly concentrated in the films. In the thick films, however, the absorbed Pd(acac)2 vapor is diffused into the inside of the film, and thus the Pd nanoparticles cannot be concentrated at such high levels as in the thin films. An alternative way to use the metal complex vapor for the incorporation of the Pd nanoparticles into a polymer film is to reduce the metal complex preliminarily dissolved in the film that is prepared by casting the solution of the polymer and the metal complex. However, in this case, a certain amount of metal complex dissolved in the film is vaporized from the film during the reduction process; therefore, highly metallized films cannot be obtained. Preparation of Highly Metallized Surface Patterns by Selective Removal of Polymer. The selective removal of the polymer film was investigated to serve highly metallized surfaces. For this purpose, we employed RIE and pyrolysis techniques. A thin PMMA film with a thickness of 100 nm was prepared on a Si wafer and was irradiated with UV light at a dose of 7 J/cm2 through a photomask in direct contact with the film. Then, the film was then exposed to Pd(acac)2 vapor for 30 min to introduce the Pd nanoparticles. The loading of metal species selectively into a polymer thin film with a defined pattern is one of the approaches used to enhance the etching selectivity against RIE for the nanofabrication of silicon substrates.39-41 Therefore, we evaluated the metallized film thus prepared as a lithography mask for RIE. Figure 4 shows the topographic surface features at the individual steps in the RIE process by SPM, where the height profiles along the lines in the images are presented below the corresponding images. Figure 4a is an image after UV irradiation, showing that the irradiated regions are slightly depressed compared to the masked regions with an average difference between the irradiated and the masked areas being about 20 nm. It is assumed that this depression is the result of the volume loss of the PMMA film due to the release of the volatiles yielded by UV irradiation. As shown in Figure 4b, the treatment by Pd(acac)2 vapor for 30 min results in a further 60 nm depression of the irradiated regions. This height reduction may be caused by the further release of the low-molecularweight species produced by UV irradiation by thermal treatment at 180 °C. In the next step, the film was treated with oxygen plasma for 30 min to remove the polymer selectively. As shown in Figure 4c, the irradiated regions turn out to be

Figure 4. SPM topography images showing the changes in the surface height variation of the binary pattern of assembled Pd nanoparticles in the RIE process. (a) A PMMA film of 100 nm thickness coated on a Si wafer was irradiated by UV light with a dose of 7 J/cm2 through a square mesh. (b) The film was exposed to Pd(acac)2 vapor for 30 min. The film was treated with (c) O2 plasma and then (d) CF4 plasma. Profiles at the bottom of the images are the height variations along the lines indicated in the corresponding images.

9356

Langmuir, Vol. 21, No. 20, 2005

Yin et al.

Figure 5. Pd 3d doublet appears in XPS spectra of the PMMA film with assembled Pd nanoparticles and the film after the removal of the PMMA films by oxygen plasma treatment and by pyrolysis at 550 °C for 30 min in an argon gas flow.

bright after the oxygen plasma treatment, indicating that the polymer around the irradiated regions was successfully removed. The height profile indicates that the irradiated regions are arranged as the protrusion of 20 nm in height with the retention of the pattern defined by the photomask. Therefore, in the PMMA film with 100 nm thickness, the metallized surface structures with binary patterns can be fabricated with 20 nm height. We also evaluated the pattern transfer onto the underlying substrate by subsequent CF4 plasma treatment. Figure 4d shows the SPM topography image and the height profile after the CF4 plasma treatment for 30 min. The metallized regions on the Si wafer work as an etching mask against the CF4 plasma, thus the bare Si wafer can be etched at a higher rate to produce surface features with significant roughness, where the height difference between the irradiated and the masked regions is increased to 700 nm from 20 nm. We have confirmed that PMMA is completely volatized at 500 °C in an inert atmosphere by TGA measurement. Therefore, pyrolysis can also remove the PMMA film selectively from the substrate while maintaining the created pattern. A metallized PMMA thin film was pyrolyzed in a tube furnace at 550 °C for 10 min in an argon flow, which resulted in a similar metallized surface pattern as shown in Figure 4c. XPS analysis was employed to investigate the valence structures of Pd. Figure 5 shows the XPS spectra of the Pd 3d region obtained from the Pd nanoparticles embedded in the PMMA film after oxygen plasma treatment and after the pyrolysis. The film with embedded Pd nanoparticles does not exhibit intense peaks because a sufficient number of Pd nanoparticles are not located on the surface. However, after the removal of the PMMA, two intense peaks corresponding to the Pd 3d5/2 and 3d3/2 doublet in the spectrum appear. The spectrum acquired from the pyrolyzed sample gives a doublet with binding energies at 335 and 340 eV that are characteristic of Pd(0), whereas the doublet in the sample treated by the oxygen plasma shifts to 337 and 344 eV, indicating that the Pd was oxidized to PdO2. Grayscale Micropatterning of Pd Nanoparticles and the Pattern Transfer to Construct 3D Surface (39) Park, M.; Harrison, C.; Chaikin, P. M.; Register, R. A.; Adamson, D. H. Science 1997, 276, 1401. (40) Spatz, J. P.; Eibeck, P.; Mossmer, S.; Moller, M.; Herzog, T.; Ziemann, P. Adv. Mater. 1998, 10, 849. (41) Lammertink, R. G. H.; Hempenius, M. A.; van den Enk, J. E.; Chan, V. Z.-H.; Thomas, E. L.; Vancso, G. J. Adv. Mater. 2000, 12, 98.

Figure 6. (a) Optical micrograph of the microlens array used to replicate a grayscale photomask. The inset is an illustration of the configuration of the lens. (b) SEM micrograph of the pattern formed in a thin PMMA film by UV photolithography through the photomask and assembled with Pd nanoparticles, and (c) SEM image of the pattern transferred onto a Si wafer by first O2 and then CF4 etchings.

Structures. Another attractive feature of our method is that the assembly of the metal nanoparticles can be tuned with continuous shades of gray by the varying the dose of UV irradiation. In other words, the density of particles distributed in PMMA films can be used to vary continuously as a function of the dose. When a gray-tone mask with graded transparency is used, the pattern thus obtained can have a graded concentration of nanoparticles with no steps. For this experiment, a microlens array (MA3007, Moritex Corp., Japan) as shown in Figure 6a was used to replicate UV light passing through a graytone photomask, thus creating a continuously varying intensity light falling on PMMA. The inset in Figure 6a illustrates the surface configuration of the lens array used for this purpose. This array is made of fused silica with a refractive index of 1.5 at a wavelength of 266 nm. A thin PMMA film with a thickness of 100 nm prepared on a Si wafer was irradiated with UV light at a dose of 7 J/cm2 through this microlens array in direct contact with it. The level of illumination of the UV light on the film was spatially varied, and thus the grayscale pattern was produced by exposure of the Pd(acac)2 vapor for 30 min as shown in Figure 6b by an SEM micrograph. The oxygen plasma and the subsequent CF4 plasma treatments yielded a 3D surface structure as shown in Figure 6c. For detailed inspections of the created surface structures, the SPM topography images and the height variations at every step of the process are shown in Figure 7. Figure 7a-d shows the surface height variations of the film after irradiation by UV light, after the exposure of the Pd(acac)2 vapor, after the oxygen plasma treatment, and after the CF4 plasma treatment, respectively. The right-hand column in the Figure shows the height profiles along the lines drawn in the corresponding images. The

Metallization of Thin Films by UV or EB Lithography

Langmuir, Vol. 21, No. 20, 2005 9357

Figure 8. SEM micrographs showing the pattern of the assembled Pd nanoparticles produced by EB lithography. (a) Low-magnification image showing the pattern created in a 200 × 200 µm2 scanning field with a dose of 125 µC/cm2. (b) Magnified image of part a showing a line produced by assembled Pd nanoparticles. (c) Backscattered image of part b.

Figure 7. Three-dimensional SPM topography images showing the changes in the surface height variation of the grayscale pattern of the assembled Pd nanoparticles in the RIE process. (a) A PMMA film of 100 nm thickness coated on a Si wafer was irradiated by UV light with the dose of 7 J/cm2 through a microlens array, (b) the film was exposed to Pd(acac)2 vapor for 30 min, and the film was treated with (c) O2 plasma and then (d) CF4 plasma. Profiles on the right side of the images are the height variations along the lines indicated in the corresponding images. The length of each SPM image corresponds to 200 µm.

surface topography of the film after the irradiation of UV light (Figure 7a) represents the projected image on the film through the microlens array. That is, the depressed region corresponds to the area where the transmission of light is relatively high. Therefore, it is recognized that the light is focused at the center region of the lenses. The tendency of the changes in the height variations throughout the process is same as the one that was observed in the process when the binary pattern was produced as shown in Figure 4. After the introduction of the Pd nanoparticles, the height variation between the dented areas and the elevated areas is a little more pronounced (Figure 7b). The treatment with O2 plasma removed PMMA selectively from the substrate, and the pattern with the inversed topographic feature was obtained (Figure 7c). The subsequent treatment with CF4 plasma etched the Si wafer at different rates because of the

gradient dispersion of the Pd nanoparticles in the film, and thus the 3D structure was produced on a Si wafer with pronounced height variation (Figure 7d). Nanoscale Assembly of Pd Nanoparticles via EB Lithography. The size of the patterned features produced via UV photolithography is on the micrometer scale. Although they could be reduced to smaller sizes by using a much finer photomask with a sophisticated irradiation system, it would still be difficult to reduce them to smaller than 100 nm because of the diffraction limit of light. PMMA has been used for a positive-type resist material for EB lithography.42,43 PMMA undergoes the scission of the main chain also by EB irradiation as by UV light. Thus, it is expected that EB would have the same effect on PMMA in enhancing the reduction power on Pd(acac)2. A thick PMMA film with 1 µm thickness on a Si wafer was used for this experiment in order to see the created pattern by SEM clearly. In a field of size 200 × 200 µm2, EB drew an image in raster scanning mode with an exposure dose of 125 µC/cm2 where several lines with different widths ranging from 500 to 100 nm were drawn onto the PMMA film. The film was then exposed to the Pd(acac)2 vapor for 30 min to introduce the Pd nanoparticles. To observe the patterned structure clearly, the film was pyrolyzed at 550 °C for 30 min in an argon gas flow to remove the PMMA film selectively. Figure 8a shows the pattern created in the scanning field by a secondary electron image in SEM. The bright region corresponds to the pattern formed by the assembly of Pd nanoparticles, where the region under the large rectangle with uniform brightness corresponds to the lines with different widths drawn by EB. Parts b and c of Figure 8 are enlarged views of the thinnest line formed with the assembled Pd nanoparticles with a width of about 100 nm by a secondary electron image and by a backscattering image, respectively. These two images from SEM represent the topographic and compositional features, respectively. The backscattering image allows us to see the individual Pd nanoparticles distributed within and around the line object. It also shows that a sufficient number of particles of uniform size form the line object by connecting each other. Undesirably, a large number of particles are (42) Chang, T. H. P.; Kern, D. P.; Kratschmer, E.; Lee, K. Y.; Luhn, H. E.; McCord, M. A.; Rishton, S. A.; Vladimirsky, Y. IBM J. Res. Dev. 1988, 32, 462. (43) Vieu, C.; Carcenac, F.; Pepin, A.; Chen, Y.; Mejias, M.; Lebib, A.; Manin-Ferlazzo, L.; Couraud, L.; Launois, H. Appl. Surf. Sci. 2000, 164, 111.

9358

Langmuir, Vol. 21, No. 20, 2005

Yin et al.

Conclusions

Figure 9. SEM images of Pd objects produced by thermal annealing at 900 °C for 30 min of the Pd assembled patterns prepared by EB lithography. (a) Pd nanowire and (b) regularly arranged Pd balls. (c) Low-magnification image of part a.

produced in the region around the line object as shown in Figure 8c. This is assumed to be caused mainly by proximity effects because such a large number of nanoparticles are not distributed in the regions outside the scanning field, which corresponds to the dark region outside the pattern shown in Figure 8a. Proximity effects are caused by forward-scattered electrons in the film and backscattered electrons from the substrate, which partially expose the area up to several micrometers from the point of impact.42 A correction for this effect can be technically achieved by introducing vector scanning mode and by using a thinner film.42,44 Figure 9 shows the SEM micrographs of the patterns of the Pd nanoparticles produced by EB lithography after the thermal annealing at 900 °C for 30 min in an argon flow. The assembled Pd nanoparticles are melted and bound into a single object to yield a metal wire (Figure 9a) and metal balls (Figure 9b) with sizes on the order of a few hundred nanometers. This suggests that, because of the nanosize effect, the melting temperature of Pd, which is 1552 °C in the bulk state, significantly decreases to 900 °C or below.45 Also, the Pd nanoparticles that are randomly distributed around the created objects combine into large particles. Figure 9c clearly shows that these randomly distributed particles are located only in the scanning area, indicating that proximity effects are the main reasons for these formations. (44) Kawano, H.; Kadowaki, Y.; Oonuki, K.; Ohta, H. Hitachi Rev. 2005, 54, 37. (45) Klabunde, K. J. Nanoscale Materials in Chemistry; WileyInterscience: New York, 2001; Chapter 8.

In this article, we present the formation of highly metallized surfaces by the assembly of Pd nanoparticles, which are synthesized with vaporized Pd(acac)2 in thin PMMA films. Photo- and EB irradiation enhance the reduction power of PMMA for Pd(acac)2 and promote the assembly of Pd nanoparticles. The obtained metallized surface patterns by UV photolithography are not limited to binary ones, which have only Pd-rich and Pd-poor regions, but also offer graded distributions of the Pd nanoparticles with no steps. This allows us to produce grayscale patterns and to construct 3D surface structures onto an underlying substrate with small feature sizes by RIE. Surfaces with 3D microstructures are required in several fields of microtechnology,46-49 such as micro-optics, integrated circuits, micro-opto-mechanical devices, and so forth. Conventionally, the fabrication of 3D structures requires several exposure steps, which is a time-consuming and costly process.50,51 Our approach can offer a convenient way to produce smoothly curved or patterned surfaces with varying heights. We also discuss how feature sizes can be reduced to the nanoscale by EB lithography. Highly metallized patterns can be created with high flexibility with the feature sizes of about 100 nm, and they can also be converted into single metal objects by annealing at a temperature that is significantly lower than the melting temperature of bulk Pd. One problem of our method is that the Pd nanoparticles are undesirably produced in unirradiated regions, which leads to the loss of pattern contrast. As long as we are using PMMA, it will not be possible to achieve perfect patterns where perfect metal-free regions are obtained in the unirradiated regions. We are therefore investigating other types of polymers to give better patterning qualities. Acknowledgment. Financial support by New Energy and Industrial Technology Development (NEDO) for the Nano-structured Polymer Project is gratefully acknowledged. The EB lithography was done using the FE-SEM/ EB system at Collabo-station II, Kyushu University. LA0511485 (46) Mu¨ther, T.; Schulze, Th.; Ju¨rgens, D.; Oberthaler, M. K.; Mlynek, J. Microelectron. Eng. 2001, 57-58, 857. (47) Yang, S.; Megens, M.; Aizenberg, J.; Wiltzius, P.; Chaikin, P. M.; Russel, W. B. Chem. Mater. 2002, 14, 2831. (48) Hirai, Y.; Harada, S.; Kikuta, H.; Tanaka, Y.; Okano, M.; Isaka, S.; Kobayasi, M. J. Vac. Sci. Technol., B 2002, 20, 2867. (49) Chen, X.; Chen, Z.; Fu, N.; Lu G.; Yang, B. Adv. Mater. 2003, 15, 1413. (50) Oikawa, M.; Iga, K.; Sanada, T.; Yamamoto, N.; Nishizawa, K. Jpn. J. Appl. Phys. 1981, 20, L296. (51) Popovic, Z. D.; Sprague, R. A.; Neville Connell, G. A. Appl. Opt. 1988, 27, 1281.