This paper presents a new method for fabricating arrayed metallic nano-structures with sub-micrometer line-widths over large patterning area sizes. It utilizes a soft mold containing arrayed surface micro-pyramids. A carbon-black photo-resist (PR) coating method is developed which can convert the soft mold into a photo-mask. This three-dimensional photo-mask is then applied for photolithographic ultraviolet (UV) patterning. In conjunction with standard metal lift-off process, arrayed metallic nano-structures are formed on glass substrates. A finite element simulation software is used to analyze the underlying mechanism of UV patterning using this new type of 3D photo-mask. The localized surface plasma resonance (LSPR) effects of the fabricated nano-structures are investigated both experimentally and theoretically. Good agreements are observed.
© 2014 Optical Society of America
Arrayed metallic nano-particles or nano-structures have recently drawn a lot of attention in researches for the possibility of manipulating their optical characteristics. For example, a two-dimensional (2D) array of nano-structures deployed on a plane can interact with an incident electromagnetic wave and induce localized surface plasma resonance (LSPR) [1–3] at certain wavelengths. Significant enhancement of absorption and scattering of optical energy can occur at these wavelengths of LSPR and therefore directly affect its transmission spectrum. The resonant frequencies and the resonance behaviors depend on the nano-structures’ material properties, sizes, shapes, array period, and geometrical arrangements, and hence are tunable.
Nano-fabrication methods are critical to the application of arrayed metallic nano-structures and their localized surface plasma resonance effects. Several methods are currently available but with some limitations. For example, electron beam lithography [4–6] is powerful in fabricating nano-structures with a feature size down to few tens of nanometers, but the patterning area size is quite limited and the cost is high. Nano-imprinting methods [7,8] can fabricate large-area nano-structures but the imprinting molds are not easily available and have a limited lifetime due to damage and contamination during imprinting processes. Plasmonic lithography [9–11] is another powerful method to fabricate nanostructures at sub-wavelength scale. However, the patterning area size is limited owing to the formation of nano-apertures on metallic film over large area using focused ion beam or electron beam lithography. Conventional photolithography methods are still the most mature and widely used technology for large-area micro/nano-patterning in different scale levels of micrometer, sub-micrometer, and nanometer. However, for the purpose of localized surface plasma resonance used in optics, the feature sizes of metallic nanostructures have to be in the order sub-micrometer or nanometer, which requires advanced and expensive photolithography equipment which is not readily accessible. In short, it is important to have a simple, low-cost, and large-area nano-fabrication method for the application of arrayed metallic nano-particles or nano-structures.
In this paper, we propose a new method which is suitable for patterning arrayed metallic nano-structures. It starts from applying the silicon bulk machining method [12,13] to create an array of inverted pyramidal micro-cavities on the surface of a (100) silicon wafer. A soft mold, which is negatively replicated from this master silicon mold by a mold process, will contain arrayed micro-pyramids on its surface. A carbon-black photo-resist (carbon-black PR), which is opaque to UV light, is spin-coated on the surface micro-pyramids of the soft mold. By carefully adjusting the spin-coating parameters, the carbon-black PR can fully cover the whole mold surface except around the tips of micro-pyramids. When using this carbon-black PR coated soft mold as a photo-mask in conventional photolithography, arrayed metallic nano-structures over large-area can be obtained in a simple and straightforward manner.
2. Experiments and analysis
To prepare the carbon-black PR coated photo-mask, a silicon master mold with arrayed and pyramid-shaped micro-cavities is first obtained based on standard silicon bulk machining method. A (100) silicon wafer is first deposited with a 100 nm thick silicon nitride (Si3N4) film using a low pressure chemical vapor deposition (LPCVD) system. Standard photolithography method and a wet etching process are subsequently applied to create arrayed hole-openings on the silicon nitride film. As shown schematically in Fig. 1(a), the hexagonal array of holes has a hole-diameter of 0.6 μm and a center-to-center pitch of 1.2 μm. The overall patterned area size of the silicon mold is 18 x 18 mm2. The nitride-patterned silicon crystal is then immersed in a 45% concentration of potassium hydroxide (KOH) solution at 65 °C for 240 seconds to create an array of pyramidal micro-cavities, as shown in Fig. 1(b). The pyramidal micro-cavities have a base length of 630 nm and a depth of 440 nm. After the silicon etching process, the nitride film is removed through hydrofluoric (HF) acid etching. This completes the preparation of a silicon master mold.
Using this silicon master mold, a carbon-black PR-coated soft photo-mask can be prepared and used for photolithography patterning as shown schematically in Fig. 2. First of all, a UV-curable polyurethane-acrylate (PUA, MINS-301RM, Minuta Technology Co., Ltd., Korea) solution is dropped and spin-coated on the silicon mold surface. The PUA layer is then cover with a 250 μm thick polyethylene terephthalate (PET, ES301445, Goodfellow Cambridge Ltd., U.K.) sheet as shown in Fig. 2(b). The ultraviolet (UV) light passing through the PET sheet can cure and solidify the PUA. After peeling off the PET sheet from the silicon mold, a soft PUA/PET mold with arrayed pyramidal surface micro-structures is formed as shown in Fig. 2(c). On top of the PUA/PET mold surface, a carbon-black photo-resist (EK410, Everlight Chemical Industrial Corporation, Taiwan) is spin-coated. The thickness of spin-coated carbon-black PR depends on the surface treatment and the spinning speed. The PUA/PET mold surface is treated with O2 plasmas at a power of 36 watts for 1 minutes such that the mold surface becomes hydrophilic. The spinning speed is adjusted by trial and error so that the carbon-black PR surface can be flush with the tips of micro-pyramids. As shown in Fig. 2(e), the obtained carbon-black PR coated soft mold can then be used as a photo-mask in UV patterning since the UV light energy can only emerge from the un-covered or barely covered tips of PUA micro-pyramids. Arrayed UV patterning over large area and with small feature size is then readily achieved.
Figure 3(a) shows the scanning electron microscope (SEM) images of a PUA/PET mold containing a hexagonal array of micro-pyramids negatively replicated from the silicon mold. The height of the micro-pyramids is 440 nm, which is the same as the depth of pyramidal micro-cavities on the silicon master mold. After being spin-coated with a carbon-black PR, Fig. 3(b) shows the SEM cross-section view of the PUA micro-pyramids covered by a carbon-black PR layer. The spin-coating speed of the carbon-black PR is 7000 rpm, which happens to coat a carbon-black PR layer with a thickness about the same as the height of micro-pyramids. When incident from the PET side, the UV light can pass through the PET and PUA materials but will be mostly absorbed and stopped by the carbon-black PR material except at the neighboring areas around the tips of PUA micro-pyramids, where the carbon-black PR is thin and some of the UV energy could emerge from the PUA tips. The UV energy emerging from the micro-pyramids’ tips can then be used for lithographically patterning of arrayed nano-structures.
To illustrate the feasibility of UV patterning using the fabricated carbon-black-PR-coated PUA/PET mold as described above, a simulation software (COMSOL Multiphysics®, COMSOL, Inc., USA) based on finite element method (FEM) is applied to analyze the UV energy distribution. A 3D model containing a PUA pyramid, the carbon-black PR material, and a 100 nm thick AZ1500 PR (AZ Electronic Materials Taiwan Co., Ltd., Taiwan) deposited on a glass substrate is created. The cross-section view of this model is shown in Fig. 4(a). The dimensions of the 3D model used here are following the experimental results shown in the SEM image of Fig. 3(b). In the simulation, the optical refraction index of the AZ1500 PR and the PUA are 1.712  and 1.49 , respectively, and the carbon-black PR has an extinction coefficient (k) of 0.14 . Matching boundary condition is applied on both the top and bottom surfaces of the model while periodic boundary conditions are applied at the vertical surfaces.
To simulate the UV energy distribution, a plane harmonic wave of UV light with a wavelength of 365 nm is incident from the top surface of the model. The electromagnetic wave (EM wave) field within the model is calculated and the normalized distribution of UV light energy is shown in Fig. 4(b). As can be seen in Fig. 4(b), most the UV energy is absorbed by the carbon-black PR material and cannot reach the AZ1500 PR layer. However, at region around the pyramid tip, some UV energy can emerge and reach the AZ1500 PR layer due to the sharp pyramidal geometry and the thinner carbon-black PR layer. When a threshold of UV energy intensity is chosen, the UV exposed area on the AZ1500 PR layer can be determined. This analysis theoretically illustrates the feasibility of using the carbon-black-PR-coated PUA/PET mold for photolithographic UV patterning.
3. Metallic arrayed nano-structures and LSPR effects
The carbon-black-PR-coated PUA/PET soft mold as prepared and described in previous section is now ready for photolithographic patterning of arrayed nano-structures. For the purpose of optical measurement, a 0.7 mm thick glass substrate is used and a 100 nm thick AZ1500 PR layer is first spin-coated on it. After soft-baking of the PR layer, the carbon-black-PR-coated PUA/PET soft mold is brought into contact with the PR/glass-substrate as shown in Fig. 2(e). Due to its flexibility, the carbon-black-PR-coated PUA/PET mold can form an intimate and conformal contact with the PR/glass-substrate. A collimated UV light source (ELS-201SA, ELS System Technology Co., Ltd., Taiwan) with variable intensity up to 20 mW/cm2 at the wavelength of 365 nm is incident from the mold. UV energy emerging from the pyramidal tips will expose the 100 nm thick AZ1500 PR layer. After PR developing, an arrayed of hole openings is formed in the PR layer. After hard-baking of the PR layer, a standard metal lift-off process is applied to complete the fabrication of metallic arrayed nano-structures.
The fabricated metallic nano-structures have the same hexagonally arrayed dot-pattern with a center-to-center distance of 1.2 μm. The diameter of the arrayed dots is variable and is depending on the UV patterning and PR developing parameters. For a UV exposure time of 6 seconds at a power of 20 mW/cm2 and a PR developing time of 15 seconds, the obtained arrayed metallic dots after lift-off have a diameter of 400 nm, as shown in Fig. 5(a). The metal material used in the lift-off process is gold (Au) with a layer thickness of 15 nm. For the same parameters and conditions but an 8 seconds UV exposure time, the average diameter of obtained metallic arrayed dots becomes 500 nm and the SEM image of obtained metallic dots are shown in Fig. 5(b).
The LSPR effects of the two arrayed metallic nano-structures formed on a glass substrate as shown in Fig. 5(a) and 5(b) are investigated by measuring their transmission coefficients using a Fourier transform infrared spectrometry (FTIR, Nicolet 6700, Thermo Fisher Scientific Inc., USA). The measured results are shown in Fig. 6(a) and 6(b) for samples shown in Fig. 5(a) and 5(b), respectively. The measurements are carried out with the wavelength ranging from 0.8 to 2.0 μm. Both samples demonstrate strong LSPR effect around the wavelength 1.5~1.7 μm by showing a significantly lower and valley-shaped optical transmission coefficient. For the arrayed nano-structures sample with a 400 nm dot-diameter, the strongest LSPR happens at the wavelength of 1.55 μm as shown in Fig. 6(a). While for the other similar arrayed nano-structures but with a 500 nm dot-diameter, the wavelength corresponding to the strongest LSPR is shifted to 1.7 μm. That is, for the hexagonally arrayed metallic dots, lager dot diameter will cause some red-shift.
To investigate the nature of the LSPR observed in Fig. 6(a) and 6(b), the FEM software COMSOL Multiphysics® is used again for simulating the electromagnetic behaviors of the arrayed nano-structures when subject to an incident plane EM wave. Figure 7(a) shows the model created by the FEM software for simulating the transmission of a normally incident plane EM wave. The theoretical transmission coefficients obtained from FEM simulations are displayed in dash-lines in Fig. 6(a) and 6(b) for arrayed metallic nano-structures with dot-diameters of 400 and 500 nm, respectively. The theoretical curves are in good agreement with their experimental counterparts. Figure 7(b) shows the magnitude of the electric field at a resonant frequency. It is quite clear that the incident EM wave is interacting with the arrayed metallic disks and induces electron oscillation inside each individual metallic disk. Therefore, the ratio of dot-diameter to the wavelength dominates the LSPR phenomenon. This explains the red-shift phenomenon observed both in the experimental and theoretical data shown in Fig. 6(a) and 6(b).
This work successfully demonstrates a new method for fabricating arrayed metallic nano-structures with small line-width, high density, and large patterning area size. It follows conventional method of photolithography but using a carbon-black-PR-coated PUA/PET soft mold with arrayed pyramidal 3D surface micro-structures as the photo-mask. This approach requires only relatively simple equipment and processes at a must lower cost, and therefore opens up a new way and possibilities of applying LSPR for real and industrial applications in optical and optoelectronic engineering.
The smallest feature size obtained in this paper is 400 nm, and hence the LSPR is in the near infrared spectrum. However, with some refinements and adjustments in the fabrication processes, it is possible for further reducing the feature sizes down to the order of 100 nm and therefore can be directly applied to LSPR in optical visible regions. The obtained metallic nano-structures obtained in this work are arrayed dots, which after subjecting to thermal annealing can turn into arrayed nano-particles as been described before [17,18]. This will further improve and enhance the LSPR effects.
This work is supported by the National Science Council (NSC) of Taiwan through project NSC102-2120-M-006-001-CC1.
References and links
1. E. Hutter and J. H. Fendler, “Exploitation of localized surface plasmon resonance,” Adv. Mater. 16(19), 1685–1706 (2004). [CrossRef]
2. F. Hallermann, C. Rockstuhl, S. Fahr, G. Seifert, S. Wackerow, H. Graener, G. V. Plessen, and F. Lederer, “On the use of localized plasmon polaritons in solar cells,” Phys. Status Solidi 205(12), 2844–2861 (2008). [CrossRef]
4. N. Felidj, J. Aubard, G. Levi, J. R. Krenn, G. Schider, A. Leitner, and F. R. Aussenegg, “Enhanced substrate-induced coupling in two-dimensional gold nanoparticle arrays,” Phys. Rev. B 66(24), 245407 (2002). [CrossRef]
5. J. Stodolka, D. Nau, M. Frommberger, C. Zanke, H. Giessen, and E. Quandt, “Fabrication of two-dimensional hybrid photonic crystals utilizing electron beam lithography,” Microelectron. Eng. 78–79, 442–447 (2005). [CrossRef]
6. Y. Chu, E. Schonbrun, T. Yang, and K. B. Crozier, “Experimental observation of narrow surface plasmon resonancesin gold nanoparticle arrays,” Appl. Phys. Lett. 93(18), 181108 (2008). [CrossRef]
7. Y. C. Lee and C. Y. Chiu, “Micro-/nano-lithography based on the contact transfer of thin film and mask embedded etching,” J. Micromech. Microeng. 18(7), 075013 (2008). [CrossRef]
9. W. Srituravanich, N. Fang, C. Sun, Q. Luo, and X. Zhang, “Plasmonic Nanolithography,” Nano Lett. 4(6), 1085–1088 (2004). [CrossRef]
11. Y. Kim, H. Jung, S. Kim, J. Jang, J. Y. Lee, and J. W. Hahn, “Accurate near-field lithography modeling and quantitative mapping of the near-field distribution of a plasmonic nanoaperture in a metal,” Opt. Express 19(20), 19296–19309 (2011). [CrossRef] [PubMed]
12. G. T. A. Kovacs, N. I. Maluf, and K. E. Petersen, “Bulk micromachining of silicon,” Proc. IEEE 86(8), 1536–1551 (1998). [CrossRef]
13. I. Barycka and I. Zubel, “Silicon anisotropic etching in KOH-isopropanol etchant,” Sens. Actuator A-Phys. 48(3), 229–238 (1995). [CrossRef]
14. R. A. Norwood and L. A. Whitney, “Rapid and accurate measurements of photoresist refractive index dispersion using the prism coupling method,” Proc. SPIE 2725, 273–280 (1996). [CrossRef]
15. http://www.minuta.co.kr/products/products_mold_template.html (Accessed February 5, 2014)
16. http://www.everlightchemical-ecbu.com/EN/product_detail.asp?seq=70 (Accessed February 5, 2014)
17. M. Bechelany, X. Maeder, J. Riesterer, J. Hankache, D. Lerose, S. Christiansen, J. Michler, and L. Philippe, “Synthesis Mechanisms of Organized Gold Nanoparticles: Influence of Annealing Temperature and Atmosphere,” Cryst. Growth Des. 10(2), 587–596 (2010). [CrossRef]
18. C. H. Chen and Y. C. Lee, “Fabrication of metallic micro/nano-particles by surface patterning and pulsed laser annealing,” Thin Solid Films 518(17), 4786–4790 (2010). [CrossRef]