Simplest Method for Creating Micropatterned Nanostructures on

Oct 7, 2011 - The fabrication of micropatterned structures on PDMS is a critical step in soft lithography, microfluidics, and many other PDMS-based ap...
5 downloads 11 Views 2MB Size
Download PDF
LETTER pubs.acs.org/Langmuir

Simplest Method for Creating Micropatterned Nanostructures on PDMS with UV Light Chang-Ying Xue, Wei Zhang, Wan Hui Stella Choo, and Kun-Lin Yang* Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576 ABSTRACT: The fabrication of micropatterned structures on PDMS is a critical step in soft lithography, microfluidics, and many other PDMS-based applications. To substitute traditional mold-casting methods, we develop a simple method to create micropatterned nanostructures on PDMS in one step. After exposing a flat PDMS surface to a UV pen lamp through a photomask (such as a TEM grid), micropatterned nanostructures can be formed readily on the PDMS surface. We also demonstrate that fabricated PDMS can be used for the microcontact printing of protein immunoglobulin (IgG) on solid surfaces. This method is probably the simplest method of creating micropatterned nanostructures on PDMS reported so far because it does not need casting, surface coating, or chemical reagents. Only a UV pen lamp and a photomask are required, and this method can be performed under ambient conditions without vacuum. We expect that this method will greatly benefit researchers who use PDMS regularly in various applications such as soft lithography and microfluidics.

’ INTRODUCTION Poly(dimethylsiloxane) (PDMS), a silicon-based polymer with good flexibility and gas permeability, has been widely used in various applications and research activites.1 More recently, PDMS with micropatterned structures on its surface was exploited for applications such as soft lithography,2 10 including microcontact printing (μCP),11 replica molding (REM),12 micromolding in capillaries (MIMIC),13 and many others.14 Thanks to its unique physical and chemical properties, micropatterned structures on PDMS can be fabricated easily by casting a liquid precursor on a silicon wafer whose surface has been patterned with photolithography. After being cured, the PDMS stamp can be peeled off of the silicon wafer easily. Although this method is very popular, making a master silicon wafer is time-consuming and expensive, and it can be damaged easily during the process. Even though the silicon master can be replaced by other materials, damage to the stamp structure can not always be avoided during the peeling process. This peeling process can also damage the micropatterned photoresist on the master, creating defects on the PDMS stamp. Currently, several solutions have been proposed to overcome these problems. For example, Nuzzo et al. developed decal transfer lithography (DTL) to fabricate micropatterned PDMS on solid substrates.15 Jalabert and coworkers fabricated a PDMS stamp with a micropatterned Au sheet on its surface. The Au pattern is released mechanically during the peel-off of the cured PDMS.16 Although these methods are useful, most micropatterned PDMS is still made following the molding process because of its simplicity and convenience. The surface treatment of PDMS with UV light has been studied and well documented.17 These studies show that UV r 2011 American Chemical Society

light can change the surface properties of PDMS such as improving the hydrophilicity and adhesion.18 In some studies, UV treatment was used to create micropatterns on flat PDMS. For example, Yeom et al. deposited a photoresist layer on flat PDMS and then fabricated 3D microstructures on the PDMS surface.19 Scharnweber and co-workers modified PDMS with vacuum UV (VUV) under ambient conditions and then used a 1:1 mixture of a 1 M aqueous solution and ethanol to develop microstructures.20 These methods can be used to make micropatterns on flat PDMS without a master mold. However, chemical reagents (as developers) are required in this process, and sometimes high vacuum is needed for these processes.21 Moreover, in some cases, micropatterns were made on the surfaces of the coating materials, not on PDMS. Unlike these previous studies, we report a one-step UV lithography method that can be used to fabricate micropatterned nanostructures on PDMS under ambient conditions without additional coating materials or chemical reagents. This method greatly simplifies the preparation of micropatterned PDMS and can also be combined with a quartz mask with chromium patterns to create complicated structures with a 1 μm feature size on PDMS.

’ MATERIALS AND METHODS Materials. TEM grids (copper, 1000 mesh) were obtained from Electron Microscopy Sciences (U.S.). All glass slides were purchased from Marienfeld (Germany). N,N-Dimethyl-N-octadecylReceived: July 30, 2011 Revised: October 4, 2011 Published: October 07, 2011 13410

dx.doi.org/10.1021/la2029824 | Langmuir 2011, 27, 13410–13414

Langmuir

LETTER

Figure 1. Microscopy image of a flat PDMS surface after UV exposure through a TEM grid (1000 mesh) for 7 min.

3-aminopropyltrimethoxysiyl chloride (DMOAP) and human IgG were purchased from Sigma-Aldrich (Singapore). Poly(dimethylsiloxane) (PDMS) stamps were prepared from Sylgard 184 (Dow Corning, Midland, MI, U.S.). A Cy3 monoreactive dye pack was purchased from GE Healthcare (U.K.). Human IgG was labeled with Cy3 following the manufacturer’s standard protocol. All aqueous solutions were prepared in deionized water with a resistance of 18.2 MΩ cm 1. Fabrication of PDMS Stamps. To prepare PDMS, the prepolymer and curing agent were first mixed in a ratio of 10:1 by mass. The precursor mixture was then cast on a clean, flat silicon wafer. The air bubbles in the precursor were removed by degassing in vacuum. After that, the stamp was cured at 100 °C for 3 h. Next, a Soxhlet device was used to extract unreacted starting materials of PDMS into ethanol. The Soxhlet extraction is used to remove unreacted monomers from PDMS because it has been reported that the cleaning step is important if PDMS is used for microcontact printing. Finally, the clean PDMS stamp was baked at 100 °C for 1 h to vaporize any ethanol trapped inside it. Direct UV Lithography on PDMS. TEM grids with square holes were first placed on a flat PDMS surface. The PDMS surface was then moved to an open glass tube equipped with a UV pen lamp (Spectronics, model 11SC-1, 254 nm) for UV exposure. The distance between the stamp and the lamp is around 1.5 cm.22 24 The power of the UV pen lamp is 4500 μW/cm2, and it provides 1.89 J/cm2 during 7 min of exposure time. In this configuration, only square regions, which were not shielded by metal bars of the TEM grid, were exposed to UV. After UV exposure, the grids were removed from the PDMS surface and the surface was ready for analysis and applications such as μCP. Preparation of a Chromium Quartz Photomask. A chromium quartz photomask was fabricated at the SERC Nanofabrication and Characterization (SNFC) facility at the Institute of Materials Research and Engineering (Singapore) following standard lithography protocols.25 The prepared chromium quartz photomask had the desired straight chromium lines (width = 1 μm) and a distance of 2 μm between every two lines.

Surface Characterization by Atomic Force Microscopy (AFM). The atomic force microscope (Nanoscope IIIa) used in this study was manufactured by Digital Instruments (U.S.). All images were collected in air by using the tapping mode of a monolithic silicon tip. The surfaces were scanned under a driving frequency of 330 ( 50 kHz with a scan rate of between 0.5 and 1.0 Hz.

Surface Characterization by Field Emission Scanning Electron Microscopy (FESEM). FESEM (JSM-6700F, JEOL, Japan) was used to examine the surface topography of UV-treated flat PDMS at an accelerating voltage of 5 kV. To make PDMS conductive, a thin layer of platinum was coated on the surface of PDMS by using an auto fine

Figure 2. AFM images of flat PDMS after 7 min of UV exposure through a TEM grid: (a) 3D topography, (b) 2D surface image, (c) phase image, and (d) depth profile along the line in part b. coater (JEOL, Tokyo, Japan) with a current intensity of 20 mA and a coating time of 40 s. Microcontact Printing of Proteins. First, the UV-treated flat PDMS surface was covered with PBS buffer containing 20 μg/mL of Cy3 human IgG. After incubation for 1 h, the protein solution was removed, and the stamp was rinsed with DI water. After being dried under a stream of nitrogen, the PDMS stamp was brought into conformal contact with a DMOAP-coated glass slide. The preparation of DMOAP-coated glass slides can be found in our previous study.26 After 2 min of contact, the stamp was peeled off of the surface, and the glass surface was dried with nitrogen for subsequent analysis. Fluorescence Microscopy. The presence of fluorescent protein Cy3 human IgG on solid surfaces was examined by using a fluorescence microscope (Eclipse E200) manufactured by Nikon (Japan). This microscope was equipped with a Cy3 filter (Chroma, U.S.). All images were captured by using a digital camera (ACT-2U, Nikon) mounted on top of the fluorescence microscope.

’ RESULTS AND DISCUSSION Formation of Nanostructures on PDMS. To fabricate micropatterned nanostructures on a flat PDMS surface, we first placed a TEM grid (1000 mesh) with square holes on the PDMS surface and then exposed it to a UV pen lamp for 7 min. In this configuration, only square regions not shielded by metal bars of the TEM grid were exposed to UV. After the UV exposure, the 13411

dx.doi.org/10.1021/la2029824 |Langmuir 2011, 27, 13410–13414

Langmuir

LETTER

Figure 4. (a) Image of the chromium photomask used to prepare the micropattern. (b) Image of a flat PDMS stamp after being exposed to UV through the photomask for 10 min. (c) Fluorescent image of protein lines printed on a DMOAP-coated glass slide by using the PDMS stamp above. (d) Fluorescent image of remaining protein lines on the PDMS stamp after μCP. The protein solution used in this experiment was 20 μg/mL of Cy3 human IgG. Figure 3. Surface topography of flat PDMS after two UV exposures through a TEM grid: (a) 2D surface image and (b) 3D topography.

TEM grid was removed from the PDMS surface. The micrograph in Figure 1 shows that square, defect-free micropatterns, which resemble the patterns on the TEM grid, were generated on the PDMS surface. This observation motivated us to investigate the surface topography further by using AFM. The image in Figure 2a shows concave squares surrounded by convex bars on the surface. These squares have dimensions of 19 μm  19 μm (same as the dimensions of the TEM grid) and a depth of about 73.9 ( 3.5 nm (Figure 2d). This result implies that UV exposure causes some shrinkage in these square areas. One possible reason is that the low-pressure mercury UV lamp emits at 185 nm, which can lead to the dissociation of oxygen molecules to form ozone.27 The generated ozone can cause the rapid oxidation of the methyl group of PDMS and can lead to the formation of new Si O Si bonds on the exposed area.28 This silica-like structure is denser and more compact, resulting in the shrinkage of PDMS. Indeed, when we examine the phase image of the UV-exposed area in Figure 2c, it is obvious that the UV-exposed areas are a different color, which is indicative of a

higher modulus. This is consistent with the proposition that a silica-like layer is formed in these areas. To understand the role of oxygen in the formation of nanostructures, we repeated the above experiment in an oxygen-free environment. Before UV exposure, the tube was filled with nitrogen to replace all air. Surprisingly, under this condition, micropatterned nanostructures cannot be formed on the surface (data not shown). This result is very different from some VUV processes in which the UV exposure was carried out in vacuum. Furthermore, we point out that although the UV pen lamp employed in this study has its main emission peak at 254 nm, other 254 nm UV lamps manufactured by UVP (U.K.) do not have the same effects on PDMS. Because the UV pen lamp is a low-pressure mercury lamp, undoubtedly it also emits UV at 185 nm. This short-wave UV can pass thorough the fused-quartz envelope and dissociate oxygen molecules to form ozone. These results, when combined, show that exposure of PDMS to a UV pen lamp can lead to the formation of micropatterned nanostructures on the surface of PDMS, and oxygen is essential in this process. However, more experiments are needed to elucidate the exact mechanism of this phenomenon. Creating Three-Dimensional Nanostructures on PDMS. We further investigated whether this method can be employed 13412

dx.doi.org/10.1021/la2029824 |Langmuir 2011, 27, 13410–13414

Langmuir to create more complex 3D microstructures through sequential UV exposure. After flat PDMS was exposed to UV through a TEM grid for 5 min, we nudged the TEM grid on the PDMS surface and continued the UV exposure for another 5 min. The result of an AFM topography image in Figure 3a shows that the PDMS surface exhibits an interlaced square micropattern after double UV exposures. Also, Figure 3b shows a two-layer, 3D nanostructure on the modified PDMS surface. The depth profile in Figure 3b suggests that the regions subjected to UV exposure twice have a depth of 102.7 ( 5.0 nm, whereas the depth in this region receives only one UV dosage (either during the first or second UV exposure) is 56.7 ( 5.4 nm or 56.3 ( 6.2 nm. This result suggests that UV modification has accumulative effects on the PDMS surface. It also indicates that complex multilayered microstructures can be prepared simply by the sequential exposure of the PDMS surface to UV under different masks. Effect of a Quartz Photomask. To investigate whether this method is general enough to fabricate micropatterned nanostructures on PDMS by using a photomask, we prepared a quartz photomask with raised, straight chromium lines (width = 1 μm) as shown in Figure 4a. Subsequently, we placed the photomask on a flat PDMS surface (with the chromium lines directly in contact with the surface) and exposed the PDMS to UV through the photomask for 10 min. Figure 4b shows the FESEM image of the UV-treated PDMS. Clearly, lines with a 1 μm width were generated on the PDMS surface. However, we noticed that the nanostructures in Figure 4b show some small defects on the surface of PDMS. This is probably caused by the presence of the quartz mask. Because UV has to penetrate the quartz photomask, some diffraction or reflection of UV might occur. This phenomenon probably results in some uneven shrinkage of PDMS, and that causes these small cracks. In contrast, when TEM grids were used, no cracks can be observed as shown in Figure 1. Despite the presence of cracks, the depth of the nanostructures shown in Figure 4b is 104.3 ( 7.0 nm (after 10 min of exposure), which is comparable to that when TEM grids were used. Because we have demonstrated that oxygen is needed in this process, there are probably pockets of air trapped in space between raised chromium lines. This air is sufficient to allow the photooxidation of PDMS when a quartz mask is used. This experiment demonstrates that a photomask can be used together with a UV pen lamp to generate micropatterned nanostructures on PDMS in one step. The possibility of using a photomask with custommade micropatterns greatly enhances the flexibility of this technique. Compared to traditional molding processes in which the curing of PDMS might damage the mold after a few runs, UV causes no damage to the photomask in our process. Therefore, the photomask can be used for a much longer period of time. Application of UV-Treated PDMS to Microcontact Printing (μCP). Because one major application of PDMS with micropatterned structures is soft lithography, we also studied whether the PDMS stamp prepared by using our method is suitable for the microcontact printing of proteins. We first dispensed PBS buffer containing 20 μg/mL of Cy3 human IgG on a UV-treated PDMS stamp. After 1 h of incubation, the stamp was rinsed with DI water and dried under nitrogen gas. Then, it was brought into conformal contact with a glass slide for 2 min. A fluorescent image of the glass slide after μCP is shown in Figure 4c. Clear, 1-μm-wide straight red lines suggest the successful transfer of proteins from the PDMS stamp to the solid surface. Meanwhile, the fluorescent image on the PDMS surface after μCP

LETTER

(Figure 4d) also shows a complementary fluorescence pattern, which indicates that proteins on the UV-exposed regions were not transferred during μCP. These results indicate that protein micropatterns can be printed on a solid surface by the UV-treated PDMS stamp with dimensions of down to 1 μm. Interestingly, considering the low aspect ratios of our PDMS stamp (h/l = 0.012 and h/d = 0.004, where h is the pillar height, l is the pillar length, and d is the roof width) in Figure 1, it is unexpected that our nanostructure does not collapse during μCP. It has been reported in the literature that to prevent the collapse of normal PDMS nanostructures the aspect ratios must be in the range of 0.5 < h/l < 5 and h/d g 0.05.29 However, roof collapse did not happen as is evident in our successful microcontact printing results. One reasonable explanation is that the UV exposure leads to the formation of a silica-like layer on the surface of PDMS, making it denser and harder. This reinforced layer prevents the collapse of these nanostructures.

’ CONCLUSIONS In summary, we demonstrated a method of fabricating micropatterned nanostructures on a flat PDMS surface with direct UV exposure. This method is probably the simplest method to date because only a UV pen lamp is needed and it can be operated under ambient conditions. By exposing a flat PDMS surface to a UV pen lamp through a photomask, a highly ordered 2D or 3D micropattern can be created on the flat PDMS surface. The photomask can be a TEM grid or a quartz disk with coated chromium patterns on its surface, although the TEM grid consistently gives better results. The UV-treated flat PDMS can be used for applications in soft lithography such as microcontact printing. Because this fabrication method is very simple and low cost, we expect that it will benefit many researchers who need to prepare PDMS regularly for soft lithography and microfluidics. ’ AUTHOR INFORMATION Corresponding Author

*Phone: (+65) 6516-6614. Fax: (+65) 6779-1936. E-mail: cheyk@ nus.edu.sg.

’ ACKNOWLEDGMENT We thank Vincent Lim Shien Fuh for his assistance with photomask fabrication and the SERC Nanofabrication and Characterization (SNFC) facility at the Institute of Materials Research and Engineering for the use of their equipment and facilities. This work was supported by research funding from the National University of Singapore and the Agency for Science, Technology and Research (A*STAR) under project number 082 101 0027. ’ REFERENCES (1) Abbasi, F.; Mirzadeh, H.; Katbab, A.-A. Polym. Int. 2001, 50, 1279–1287. (2) Wu, H.; Odom, T. W.; Chiu, D. T.; Whitesides, G. M. J. Am. Chem. Soc. 2003, 125, 554–559. (3) Quake, S. R.; Scherer, A. Science 2000, 290, 1536–1540. (4) Rogers, J. A.; Nuzzo, R. G. Mater. Today 2005, 8, 50–56. (5) Tan, H.; Huang, S.; Yang, K.-L. Langmuir 2007, 23, 8607–8613. (6) Chen, C. H.; Yang, K.-L. Analyst 2011, 136, 733–739. (7) Ziolkowska, K.; Kwapiszewski, R.; Brzozka, Z. New J. Chem. 2011, 35, 979–990. 13413

dx.doi.org/10.1021/la2029824 |Langmuir 2011, 27, 13410–13414

Langmuir

LETTER

(8) Kartalov, E. P.; Anderson, W. F.; Scherer, A. J. Nanosci. Nanotechnol. 2006, 6, 2265–2277. (9) Singh, P. K.; Sharma, V.; Tanwar, V. K.; Jain, S. C. J. Optoelectron Adv. Mater. 2007, 9, 127–133. (10) Michel, B.; Bernard, A.; Bietsch, A.; Delamarche, E.; Geissler, M.; Juncker, D.; Kind, H.; Renault, J. P.; Rothuizen, H.; Schmid, H.; Schmidt-Winkel, P.; Stutz, R.; Wolf, H. Chimia 2002, 56, 527–542. (11) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498–1511. (12) Xia, Y.; Kim, E.; Zhao, X.-M.; Rogers, J. A.; Prentiss, M.; Whitesides, G. M. Science 1996, 273, 347–349. (13) Kim, E.; Xia, Y.; Whitesides, G. M. Nature 1995, 376, 581–584. (14) Xia, Y.; Whitesides, G. M. Annu. Rev. Mater. Sci. 1998, 28, 153–158. (15) Childs, W. R.; Nuzzo, R. G. J. Am. Chem. Soc. 2002, 124, 13583– 13596. (16) Jalabert, L.; Bottier, C.; Kumemura, M.; Fujita, H. Microelectron. Eng. 2010, 87, 1431–1434. (17) Ye, H.; Gu, Z.; Gracias, D. H. Langmuir 2006, 22, 1863–1868. (18) Chen, C.-H.; Yang, K.-L. Langmuir 2011, 27, 5427–5432. (19) Yeom, J.; Shannon, M. A. Adv. Funct. Mater. 2009, 19, 1–7. (20) Scharnweber, T.; Truckenmuller, R.; Schneider, A. M.; Welle, A.; Reinhardta, M.; Giselbrecht, S. Lab Chip 2011, 11, 1368–1371. (21) Wu, Z.; Yan, H.; Chen, H.; Huang, H. Appl. Surf. Sci. 2009, 255, 4702–4704. (22) Xue, C.-Y.; Chin, S.-Y.; Khan, S. A.; Yang, K.-L. Langmuir 2010, 26, 3739–3743. (23) Zhang, W.; Xue, C.-Y.; Yang, K.-L. J. Colloid Interface Sci. 2011, 353, 143–148. (24) Xue, C.-Y.; Yang, K.-L. J. Colloid Interface Sci. 2010, 344, 48–53. (25) Larsson, A.; Du, C.-X.; Liedberg, B. Biomacromolecules 2007, 8, 3511–3518. (26) Xue, C.-Y.; Yang, K.-L. Langmuir 2008, 24, 563–567. (27) Muisiner, R. J.; Koberstein, J. T. Polym. Mater. Sci. Eng. 1997, 77, 653. (28) Schnyder, B.; Lippert, T.; Kotz, R.; Wokaun, A.; Graubner, V.-M.; Nuyken, O. Surf. Sci. 2003, 532 535, 1067–1071. (29) Qin, D.; Xia, Y.; Whitesides, G. M. Nat. Protoc. 2010, 5, 491–502.

13414

dx.doi.org/10.1021/la2029824 |Langmuir 2011, 27, 13410–13414