pubs.acs.org/Langmuir © 2010 American Chemical Society
Photocurable Silsesquioxane-Based Formulations as Versatile Resins for Nanoimprint Lithography Bong Kuk Lee,†,‡ Nam-Goo Cha,† Lan-Young Hong,§ Dong-Pyo Kim,§ Hidekazu Tanaka,† Hea Yeon Lee,*,†,^ and Tomoji Kawai†,^ † The Institute of Scientific and Industrial Research, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan, ‡Bionanotechnology Research Center, Korea Research Institute of Bioscience & Biotechnology, 111 Gwahangno, Yuseong-gu, Daejeon 305-806, Korea, §Department of Fine Chemical Engineering and Chemistry, 220 Kung-dong, Yusung-gu, Daejeon 305-764, Korea, and ^Division of Quantum Phases & Devices, School of Physics, Konkuk University, Seoul 143-701, Korea
Received June 21, 2010. Revised Manuscript Received August 6, 2010 Methacrylate octafunctionalized silsesquioxane (SSQMA) was shown to be an ideal material with high performance for ultraviolet (UV)-based nanoimprint lithography (NIL). The total viscosity of SSQMA-based formulations was adjusted to between 0.8 and 50 cP by incorporating low-viscosity acrylic additives, making the formulations suitable for UV-based NIL. The cured SSQMA-based formulations showed numerous desirable characteristics, including low volumetric shrinkage (4%), high Young’s modulus (2.445-4.272 GPa), high resistance to oxygen plasma, high transparency to UV light, and high resistance to organic/aqueous media, as a functional imprint material for UVbased NIL and step-and-flash imprint lithography (SFIL). Using both techniques, the SSQMA-based formulations were easily transferred to relief structures with excellent imprint fidelity and minimal residual thickness. Formulations containing 50% SSQMA (wt %) were able to reproduce high-aspect-ratio nanostructures with aspect ratios as high as 4.5 using bilayer SFIL. Transparent rigiflex molds and hard replica molds with sub-50-nm size features were reproducibly duplicated by using UV-NIL with the SSQMA-based resin. Nanostructures with feature sizes down to 50 nm were successfully reproduced using these molds in both UV- and thermal-NIL processes. After repeating 20 imprinting cycles at relatively high temperature and pressure, no detectable collapse or contamination on the replica surface was observed. These properties of the SSQMA-based resins make them suitable as inexpensive and convenient components in all NIL processes that are based on physical contact.
1. Introduction The ability to fabricate nanostructures with new and functional materials is essential to modern science and technology.1,2 Conventional methods, such as photolithography, atomic force microscopy (AFM) lithography, and electron-beam lithography, can be used for nanofabrication,3 but they are expensive, slow, and difficult to apply in many applications, such as a nonflat surface. In contrast, unconventional methods such as soft lithography4,5 and nanoimprint lithography (NIL)6,7 are simple and costeffective. The NIL techniques include thermal NIL (T-NIL),6,7 ultraviolet (UV)-NIL,8 and step-and-flash imprint lithography (SFIL).9,10 UV-NIL and SFIL are especially suitable for high-density, highthroughput, and low-cost approaches to nanofabrication. These UV-based NILs can be performed at room temperature with low *Corresponding author: e-mail
[email protected]; Tel þ81-66879-4280; Fax þ81-6-6875-4283.
(1) Quake, S. R.; Scherer, A. Science 2000, 290, 1536. (2) Koch, S. W.; Knorr, A. Science 2001, 293, 2217. (3) Gates, B. D.; Xu, Q.; Stewart, M.; Ryan, D.; Willson, C. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1171. (4) Xia, Y.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 550. (5) Xia, Y.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. Chem. Rev. 1999, 99, 1823. (6) Chou, S. Y.; Krauss, P. R.; Renstrom, P. J. Appl. Phys. Lett. 1995, 67, 3114. (7) Chou, S. Y.; Krauss, P. R.; Renstrom, J. P. Science 1996, 272, 85. (8) Haisma, J.; Verheijen, M.; Vandenheuvel, K.; Vandenberg, J. J. Vac. Sci. Technol. B 1996, 14, 4124. (9) Colburn, M.; Johnson, S.; Stewart, M.; Damle, S.; Bailey, T. C.; Choi, B.; Wedlake, M.; Michaelson, T.; Sreenivasan, S. V.; Ekerdt, J. Proc. SPIE-Int. Soc. Opt. Eng. 1999, 3676, 379. (10) Ruchhoeft, P.; Colburn, M.; Choi, B.; Nounu, H.; Johnson, S.; Bailey, T.; Damle, S.; Stewart, M.; Ekerdt, J.; Sreenivasan, S. V. J. Vac. Sci. Technol. B 1999, 17, 2965.
Langmuir 2010, 26(18), 14915–14922
imprinting pressure in a very short time using a photocurable precursor with low viscosity.9,10 NIL processes have demonstrated potential utility in biological applications11-14 as well as in the fabrication of epitaxial nanodot arrays of transition-metal oxides,15 semiconductor devices,16 microfluidic devices,17 optical components,18 photonic devices,19 and replica mold components.20,21 Photosetting in UV-NIL resists is realized by a free-radical or cationic polymerization. In the case of cationic polymerization, the relatively long lifetime of the cationic catalysts, as well as the acidic nature of the cationic initiators, is problematic. Thus, freeradical curing of (meth)acrylated oligomers or polymers is generally used for UV-NIL resists, although this method is generally more sensitive to molecular oxygen. On the other hand, the resists (11) Hoff, J. D.; Cheng, L. J.; Meyh€ofer, E.; Guo, L. J.; Hunt, A. J. Nano Lett. 2004, 4, 853. (12) Hu, W.; Yim, E. K. F.; Reano, R. M.; Leong, K. W.; Pang, S. W. J. Vac. Sci. Technol. B 2005, 23, 2984. (13) Lee, B. K.; Lee, H. Y.; Kim, P.; Suh, K. Y.; Seo, J. H.; Cha, H. J.; Kawai, T. Small 2008, 4, 342. (14) Lee, B. K.; Lee, H. Y.; Kim, P.; Suh, K. Y.; Kawai, T. Lab Chip 2009, 9, 132. (15) Suzuki, N.; Tanaka, H.; Yamanaka, S.; Kanai, M.; Lee, B. K.; Lee, H. Y.; Kawai, T. Small 2008, 4, 1661. (16) Zhang, W.; Chou, S. Y. Appl. Phys. Lett. 2003, 83, 1632. (17) Cao, H.; Yu, Z.; Wang, J.; Tegenfeldt, J. O.; Austin, R. H.; Chen, E.; Wu, W.; Chou, S. Y. Appl. Phys. Lett. 2002, 81, 174. (18) Seekamp, J.; Zankovych, S.; Helfer, A. H.; Maury, P.; Sotomayor Torres, C. M.; B€ottger, G.; Liguda, C.; Eich, M.; Heidari, B.; Montelius, L.; Ahopelto, J. Nanotechnology 2002, 13, 581. (19) Kee, C.-S.; Han, S.-P.; Yoon, K. B.; Choi, C.-G.; Sung, H. K.; Oh, S. S.; Park, H. Y.; Park, S.; Schift, H. Appl. Phys. Lett. 2005, 86, 051101. (20) Ge, H.; Wu, W.; Li, Z.; Jung, G.-Y.; Olynick, D.; Chen, Y.; Liddle, J. A.; Wang, S.-Y.; Williams, R. S. Nano Lett. 2005, 5, 179. (21) Choi, S.-J.; Yoo, P. J.; Baek, S. J.; Kim, T. W.; Lee, H. H. J. Am. Chem. Soc. 2004, 126, 7744.
Published on Web 08/23/2010
DOI: 10.1021/la1025119
14915
Article
for basic SFIL are generally composed of a mixture of siliconcontaining monomers, cross-linkers, photoinitiators, and reactive diluents for low viscosity.9,10,22-26 Also, NIL processes can be used to directly imprint functional materials. In such applications, the imprint materials are usually designed to have a particular characteristic, such as high electrical insulation in semiconductor devices,27 high transparency in optics,18 high mechanical strength and resilience in replica mold components,20,21 high solvent resistance in microfluidic devices,28 and antibiofouling properties in biological applications.11,13,14 As there is no currently available resin that provides all of these characteristics, future applications for UV-NIL will require the development of new, low-viscosity, photocurable materials. Silsesquioxane (SSQ)-based hybrid materials have attracted considerable interest as functional materials. SSQ materials have a compact and hybrid structure, composed of an inorganic core of silicon and oxygen (SiO1.5)n (n = 8, 10, and 12), surrounded by nonreactive or reactive polymerizable organic ligands.27,29-33 The incorporation of SSQ cages into polymeric materials often results in substantial improvements in thermal stability,29,30 mechanical strength,29,32 glass transition temperature,34 and insulating properties.29,35 However, relatively few reports have described the successful fabrication of SSQ-based nanostructures using NIL, despite the numerous advantages mentioned above. Hydrogen silsesquioxane (HSQ) and spin-on-glass (SOG) compounds have been used as imprint resist materials;36,37 however, the high viscosity of HSQ and SOG requires a very high imprinting pressure, thereby limiting the utility of these materials. Recently, SSQ functionalized with both methacrylate and benzocyclobutene was directly imprinted for use as a dielectric material for dual damascene processing by SFIL.27 These results suggest the possibility of using photocurable, monomeric SSQ, which exhibits a relatively low viscosity, as a resist material in SFIL, although the viscosity of functionalized SSQ materials is at least 10-fold higher than the 20-cP viscosity required for the droplet ejection of SFIL resists using inkjet printing techniques.23 The aim of this study was to develop an SSQ-based resin with versatile functionalities for use as NIL resists and to demonstrate the material’s ability to perform in a variety of nanoimprint lithographic technologies. We describe here our best efforts to (22) Colburn, M.; Suez, I.; Choi, B. J.; Meissl, M.; Bailey, T.; Sreenivasan, S. V.; Ekerdt, J. G.; Willson, C. G. J. Vac. Sci. Technol. B 2001, 19, 2685. (23) Palmieri, F.; Adams, J.; Long, B.; Heath, W.; Tsiartas, P.; Willson, C. G. ACS Nano 2007, 1, 307. (24) Kim, E. K.; Stacey, N. A.; Smith, B. J.; Dickey, M. D.; Johnson, S. C.; Trinque, B. C.; Willson, C. G. J. Vac. Sci. Technol. B 2004, 22, 131. (25) Kim, E. K.; Stewart, M. D.; Wu, K.; Palmieri, F. L.; Dickey, M. D.; Ekerdt, J. G.; Willson, C. G. J. Vac. Sci. Technol. B 2005, 23, 2967. (26) Johnson, S.; Burns, R.; Kim, E. K.; Dickey, M.; Schmid, G.; Meiring, J.; Burns, S.; Willson, C. G.; Convey, D.; Wei, Y.; Fejes, P.; Gehoski, K.; Mancini, D.; Nordquist, K.; Dauksher, W. J.; Resnick, D. J. J. Vac. Sci. Technol. B 2005, 23, 2553. (27) Schmid, G. M.; Stewart, M. D.; Wetzel, J.; Palmieri, F.; Hao, J. J.; Nishimura, Y.; Jen, K.; Kim, E. K.; Resnick, D. J.; Liddle, J. A.; Willson, C. G. J. Vac. Sci. Technol. B 2006, 24, 1283. (28) Rolland, J. P.; Van Dam, R. M.; Schorzman, D. A.; Quake, S. R.; DeSimone, J. M. J. Am. Chem. Soc. 2004, 126, 2322. (29) Liu, Y.-L.; Tseng, M.-C.; Fangchiang, M.-H. J. Polym. Sci., Polym. Chem. 2008, 46, 5157. (30) Schwab, J. J.; Lichtenhan, J. D. Appl. Organomet. Chem. 1998, 12, 707. (31) Lichtenhan, J. D. Comments Inorg. Chem. 1995, 17, 115. (32) Bizet, S.; Galy, J.; Gerard, J.-F. Macromolecules 2006, 39, 2574. (33) Xu, Y.; Zhu, X.; Yang, S. ACS Nano 2009, 3, 3251. (34) Abad, M. J.; Barral, L.; Fasce, D. P.; Williams, R. J. J. Macromolecules 2003, 36, 3128. (35) Su, R. Q.; M€uller, T. E.; Prochazka, J.; Lercher, J. A. Adv. Mater. 2002, 14, 1369. (36) Matsui, S.; Igaku, Y.; Ishigaki, H.; Fujita, J.; Ishida, M.; Ochiai, Y.; Komuro, M.; Hiroshima, H. J. Vac. Sci. Technol. B 2001, 19, 2801. (37) Igaku, Y.; Matsui, S.; Ishigaki, H.; Fujita, J.-I.; Ishida, M.; Ochiai, Y.; Namatsu, H.; Komuro, M.; Hiroshima, H. Jpn. J. Appl. Phys. 2002, 41, 4198.
14916 DOI: 10.1021/la1025119
Lee et al.
prepare a photocurable, SSQ-based functional resist with enhanced properties, including low viscosity, high modulus, low volumetric shrinkage, high O2-etching resistance, high transparency to UV, and high resistance to organic solvents. Among the various SSQ materials, methacrylate octafunctionalized SSQ (SSQMA) was selected due to its relatively low viscosity, commercial availability, high miscibility with acrylic monomers, and potential for rapid polymerization via free-radical propagation. To obtain low-viscosity SSQMA-based resists, SSQMA was mixed with a methacrylated cross-linker and/or reactive diluent. To evaluate the properties of the SSQMA-based formulations, we investigated the viscosity, mechanical properties, volumetric shrinkage, O2-etching resistance, transparency to UV, and solvent resistance to identify desirable characteristics in the materials that can be utilized for UV-based NILs. We also demonstrated a variety of uses for the SSQMA-based formulations, such as their use as an O2-etching barrier during fabrication of high-aspectratio (HAR) structures and as components of replica molds in both UV-NIL and T-NIL with sub-50-nm features.
2. Experimental Methods Materials. SSQMA monomer was purchased from Hybrid Plastics (Fountain Valley, CA). Methyl methacrylate (MMA), tert-butyl methacrylate (tBMA), ethylene glycol dimethacrylate (EGDMA), tri(ethylene glycol) dimethacrylate (TEGDMA), 2,20 -dimethoxy-2-phenylacetophenone (DMPA), 3-(trimethoxysilyl)propyl methacrylate (TMSPM), and trichloro(1H,1H,2H,2H-perfluorooctyl)silane (PFOS) were purchased from SigmaAldrich (St. Louis, MO). NXR-3030 resists were provided by Nanonex (Monmouth Junction, NJ). PAK-01 was purchased from Toyo Gosei (Chiba, Japan). Poly(methyl methacrylate) (PMMA) was purchased from MicroChem (Newton, MA). The master molds with positive (NIM-100D), negative (NIM-100H), and line-and-space (NIM-50 L) features were supplied by NTTAT Corp. (Tokyo, Japan). Measurements of Mechanical and Physical Properties. The viscosities of SSQ-based formulations were determined at 25 C using a Brookfield viscometer, model DV-II Pro (Brookfield Engineering Laboratories Inc., Stoughton, MA).38 Measurements were acquired on 0.5-0.7-mL samples with a CPE-51 spindle at a rotation speed of 5-30 rpm. The Young’s moduli of coated resins were measured at room temperature with a commercial nanoindentation system (Nanoindenter XP; MTS Nano Instruments, Oak Ridge, TN). To remove the influence of the substrate, SSQMA-based formulations containing 2% DMPA (wt %) were deposited as a film ∼10 μm thick onto SiO2 wafers. The formulations were cured at 365 nm (1000 mJ cm-2 UV dose) with a UV lamp (Toscure251; Toshiba, Tokyo, Japan). To minimize position-dependent factors, the data presented are statistical mean values calculated from at least 10 measurements.38 Poisson’s ratio of all samples was set to 0.35, which is the value for HSQ.39 UV-Based NIL Processes. UV-NIL and SFIL were used to fabricate relief nanostructures in SSQMA-based resists. UVozone-cleaned SiO2 substrates were modified with 10 mM TMSPM diluted in dehydrated toluene (Sigma-Aldrich) for 2 h to enhance adhesion with the resist. After coupling, samples were rinsed thoroughly with anhydrous toluene and acetone and dried under a stream of N2 gas. For UV-NIL, 10% SSQMA (wt %), diluted in dehydrated toluene containing 2% DMPA (wt %) as a photoinitiator (wt %), was spin-coated onto a TMSPM-modified SiO2 substrate to form a 500-nm film. The sample was then prebaked on a hot plate at 70 C for 3 min to evaporate the solvent. For SFIL, the SSQMA-based resist solutions were dispensed as (38) Lee, B. K.; Hong, L.-Y.; Lee, H. Y.; Kim, D.-P.; Kawai, T. Langmuir 2009, 25, 11768. (39) Wu, W. L.; Liou, H. C. Thin Solid Films 1998, 312, 73.
Langmuir 2010, 26(18), 14915–14922
Lee et al. droplets onto TMSPM-modified substrates. Imprinting was performed using a nanoimprinter system (NM-401; Meisyo Kiko, Hyogo, Japan) equipped with a UV lamp (Toscure251; Toshiba). Before the imprinting process, the master molds (NIM-100D and NIM-100H) were modified with PFOS as a release agent under vacuum at 150 C for 1 h, followed by the UV-ozone treatment for 30 min with an ozone cleaner (NL-UV253; Nippon Laser Denshi, Japan). For UV-based NILs, the mold was pressed at an imprint pressure of 0.2-5 MPa at room temperature for 20 s under vacuum. The SSQMA-based resists were cured at 365 nm (200 mJ cm-2 UV dose) for ∼30-35 s, while maintaining the imprinting pressure. The mold was then detached from the substrate. The residual layer thicknesses were measured with an Alpha-Step 500 surface profiler (Tencor Instruments Inc., Mountain View, CA) and AFM, after scratching the imprinted surface with a sharp needle. To fabricate a high-aspect-ratio feature, a 450-700-nm film of Nanonex NXR-3030 resist was spin-coated onto a SiO2 substrate and prebaked on a hot plate at 70 C for 10 min to evaporate the solvent. Subsequently, SFIL was performed at 5 MPa after dispensing the SSQMA-based liquid as droplets onto the NXR-3030 resist. Prior to bilayer-type SFIL, the etching rates of the NXR-3030 resist, SSQMA, and the 50 wt % SSQMA-based resists were measured with a SAMCO reactive ion etching (RIE) system (RIE-10NR; Samco Inc., Tokyo, Japan). The gas flow rate, gas pressure, and rf power were 20 sccm, 2.0 Pa, and 50 W, respectively, for both CF4 and O2 RIE. The etching rates of the samples are summarized in Table S1 (see Supporting Information).
Fabrication of Replica Molds Using UV-Based NIL.
Approximately 1 μL of SSQMA-based resist was drop-dispensed onto a silicon master. A flexible and transparent support made of poly(ethylene terephthalate) (PET) was carefully placed on top of the surface. The PET film (Lumirror; U-34) used in this study was surface-modified with polyurethane of ca. 1 μm thickness to increase adhesion with the acrylate-containing monomer (Toray, Chiba, Japan). The support was pressed for 20 s at an imprint pressure of 0.2 MPa at room temperature under vacuum and subsequently cured at 365 nm (200 mJ cm-2 UV dose) for ∼35 s while maintaining the imprinting pressure. After UV curing, the SSQMA-based replica mold was peeled from the master using sharp tweezers. The preparation was complete after trimming the edges of the first replica mold. To use this replica as an NIL stamp, PFOS was modified to the surface of the replica mold under vacuum at 80 C for 1 h, after the O2 plasma treatment at 10 Pa for 20 s using a SAMCO RIE system. NIL Processes Using Replica Molds. The rigiflex replica mold was used as a stamp for the UV-NIL process described above. PAK-01 was used as a photocurable resist. This resist was imprinted at a pressure of 10 MPa at room temperature for 20 s under vacuum and subsequently cured at 365 nm (200 mJ cm-2 UV dose) for ∼35 s while maintaining the imprinting pressure. For T-NIL, the silicon wafer was cleaned with a UV-ozone cleaner, spin-coated with a thin film of PMMA, and then baked at 80 C for 10 min to evaporate the solvent. PMMA was imprinted with the replica mold at 160 C and 10 MPa for 5 min. After cooling to room temperature, the replica mold was separated from the substrate. Observation of Pattern Morphology. Patterned nanostructures were imaged with a Digital Instruments NanoScope III atomic force microscope (Veeco Instruments, Woodbury, NY) in tapping mode in air at ambient temperature. The scan rate was 0.8 Hz, and 256 lines were scanned per sample. Tapping mode cantilevers (NCH-10 V) with a 38 N m-1 spring constant and a radius of curvature of 5-10 nm were obtained from Veeco. The data were processed using SPIP V3.3.7.0 software (Image Metrology, Lyngby, Denmark). Field emission scanning electron microscopy (FE-SEM, S-4300 type; Hitachi Co., Tokyo, Japan) was also used to observe the surface morphology of the master and imprinted patterns. To prevent charging, the samples were Langmuir 2010, 26(18), 14915–14922
Article
Figure 1. Molecular structures of the photocurable components used for UV-NIL and SFIL. coated with a 10-nm gold layer prior to analysis using Quick Coater SC-701HMC (Sanyu Electron Co. Ltd., Tokyo, Japan).
3. Results and Discussion 3.1. Formulations of SSQMA Resist and Their Characterization. Although SSQMA has a low viscosity relative to other SSQ monomers,27 as received from Hybrid Plastics it has too high a viscosity (1800 cP) to use for SFIL resists. In this study, mono- and/or difunctional acrylic monomers were incorporated into the SSQMA resins to control the total viscosity and mechanical properties upon curing. The components used in this study are shown in Figure 1. To begin, the viscosities of the SSQMA/acrylic mixtures containing 2% DMPA (wt %) as a photoinitiator were investigated. The total viscosity depended on the intrinsic viscosity of the additive and the relative concentration (wt %) of SSQMA (see Supporting Information, Table S2). Interestingly, the viscosities of the resins were drastically reduced from 1800 cP to between 0.8 and 50 cP by adding low-viscosity acrylic monomers. These results suggest that all of the SSQMA-containing formulations listed in Table S2 are suitable as UV-NIL resists, and the formulations with viscosities 2000 times higher than that of Sylgard 184 PDMS (1.8 MPa).41 It implies that the imprinted sub-50-nm feature structures with these SSQMA-based formulations would be rigid enough to resist failures, making these formulations appropriate for UV-based NILs.23-25 3.2. Application of SSQMA-Based Formulations as UV Nanoimprint Resins. 3.2.1. UV-NIL and SFIL. The applicability of the SSQMA monomer and the SSQMA/acrylic mixtures as UV-based NIL resists was investigated. The nanostructures of SSQMA and SSQMA/acrylic mixture were fabricated by UV-NIL (Figure 3a) and SFIL (Figure 3d), respectively. Figures 3b and 3c show atomic force microscopy (AFM) and field emission scanning electron microscopy (FE-SEM) images of the negative quartz master and the imprinted 100-nm pure SSQMA patterns, respectively. The SSQMA monomer was
Figure 2. Young’s modulus, as a function of contact depth, of various cured SSQMA/acrylic compositions.
Lee et al.
reproducibly duplicated over a large area with excellent uniformity on a SiO2 substrate by UV-NIL. The mean surface roughness of the imprinted resists was less than 1 nm. The average height of the imprinted features was 178 nm compared to the 185 nm height of the master features. Photopolymerization-induced shrinkage typically results in a decrease of feature height rather than changes in the sidewall angle of nanostructured objects.26 If this effect is taken into account, then the shrinkage ratio of the pure SSQMA was ∼4% under UV-NIL conditions. Figures 3e and 3f show AFM and SEM images of the positive quartz master and the 100-nm pattern of the 50:50 wt % SSQMA/ EGDMA mixture, respectively. The SSQMA/EGDMA mixture was easily transferred to the relief structures over a large area without defects. The volumetric shrinkage of the SSQMA-based resins was determined from the average height of the imprinted resin patterns.26 Note that all of the SSQMA/acrylic mixtures listed in Table S2 exhibited, statistically, the same degree of shrinkage (∼4%). These results are encouraging because a cureinduced shrinkage has been reported to create stress in the mold and/or the molded material, although it can be potentially beneficial for mold release.42 3.2.2. Residual Layer Thickness. One of the key process requirements in SFIL is a uniform and thin residual layer. If the residual layer is uniform and as thin as possible, the time for postimprint plasma treatment is reduced. To evaluate the suitability of SSQMA-based formulations for the SFIL process, the residual layer thickness and the uniformity of the formulations were examined while varying the imprinting pressure. Figure 4a shows the residual thickness of imprinted SSQMAbased resists as a function of imprinting pressure. At low imprinting pressures, a high-viscosity formulation yielded a higher residual thickness with slightly less homogeneous uniformity than did a low-viscosity formulation. The residual layer thickness of the SSQMA-based formulations decreased with increasing pressure. At high imprinting pressure (5 MPa), the residual layer thickness for all samples was