Azo Polymer Microspherical Cap Array: Soft-Lithographic Fabrication

Sep 20, 2007 - In this work, azo polymer microspherical cap arrays possessing unique photoprocessible properties have been fabricated through a ...
0 downloads 0 Views 968KB Size
Download PDF
11266

Langmuir 2007, 23, 11266-11272

Azo Polymer Microspherical Cap Array: Soft-Lithographic Fabrication and Photoinduced Shape Deformation Behavior Bin Liu, Yaning He, Pengwei Fan, and Xiaogong Wang* Department of Chemical Engineering, Laboratory for AdVanced Materials, Tsinghua UniVersity, Beijing, 100084, P. R. China ReceiVed June 4, 2007. In Final Form: July 17, 2007 In this work, azo polymer microspherical cap arrays possessing unique photoprocessible properties have been fabricated through a soft-lithographic contact printing approach. In the process, hexagonal polystyrene (PS) colloidal arrays, obtained by the vertical deposition method, were used as masters. Poly(dimethylsiloxane) (PDMS) stamps with aligned hemisphere air voids on the surfaces were obtained by casting the precursor against the colloidal arrays. By using the stamps and a solution of an epoxy-based azo polymer (BP-AZ-CA) as “ink”, the microspherical cap arrays were fabricated by pressing the “inked” surfaces against substrates. Uniform 2D arrays of the submicrometer spherical caps could be obtained on the substrates after peeling off the stamps and drying. The characteristic sizes of the arrays depended on some adjustable features, such as the diameters of PS spheres and concentrations of the “inks” used in the process. After exposure to a linearly polarized Ar+ laser single beam, the spherical caps could be stretched along the polarization direction, and the arrays were consequently transformed into ellipsoidal cap arrays. Upon irradiation of interfering p-polarized Ar+ laser beams, only the spherical caps in the bright fringes were deformed by the light irradiation, which resulted in more complicated surface relief patterns. The observation gives another well-defined example of the photoinduced mass migration in the submicrometer scale. The approach can potentially be applied to fabrication of microlens arrays with different converging rate in two directions.

Introduction Polymers containing aromatic azo chromophores (azo polymers for short) have been extensively investigated in recent years for their interesting properties and many potential applications.1-4 One of the most important photoresponsive properties, which has been well documented but still not been fully understood, is the surface-relief-grating (SRG) formation observed on films of azo polymers.1,3 When azo polymer films are exposed to an interference pattern of coherent Ar+ laser beams, SRGs can be formed on the film surfaces at temperatures well below the glasstransition temperatures (Tgs) of the polymers. The formed surface patterns are stable below the Tgs and can be erased by heating samples to a temperature above their Tgs or upon irradiation with a circularly polarized laser beam. SRG formation is neither a thermally driven process nor ablation in the irradiation regions at the low or middle light intensity.1,3 SRG formation is closely related to the repeated trans-cis-trans isomerization cycles of the azo chromophores caused by light irradiation.5,6 SRG * To whom correspondence should be addressed. E-mail: [email protected]. (1) (a) Kumar, G. S.; Nechers, D. C. Chem. ReV. 1989, 89, 1915. (b) Delaire, J. A.; Nakatani, K. Chem. ReV. 2000, 100, 1817. (c) Natansohn, A.; Rochon, P. Chem. ReV. 2002, 102, 4139. (2) (a) Todorov, T.; Nikolova, L.; Tomova, N. Appl. Opt. 1984, 23, 4309. (b) Ikeda, T.; Horiuchi, S.; Karanjit, D.; Kurihara, S.; Tazuke, S. Macromolecules 1990, 23, 43. (3) (a) Rochon, P.; Batalla, E.; Natansohn, A. Appl. Phys. Lett. 1995, 66, 136. (b) Kim, D. Y.; Tripathy, S. K.; Li, L.; Kumar, J. Appl. Phys. Lett. 1995, 66, 1166. (4) (a) Finkelmann, H.; Nishikawa, E.; Pereira, G. G.; Warmer, M. Phys. ReV. Lett. 2001, 87, 015501. (b) Li, M. H.; Keller, P.; Li, B.; Wang, X. G.; Brunet, M. AdV. Mater. 2003, 15, 569. (c) Yu, Y. L.; Nakano, M.; Ikeda, T. Nature 2003, 425, 145. (d) Camacho-Lopez, M.; Finkelmann, H.; Palffy-Muhoray, P.; Shelley, M. Nat. Mater. 2004, 3, 307. (e) Ikeda, T.; Mamiya, J. I.; Yu, Y. L. Angew. Chem., Int. Ed. 2007, 46, 506. (5) (a) Barrett, C. J.; Rochon, P.; Natansohn, A. J. Phys. Chem. 1996, 100, 8836. (b) Barratt, C. J.; Rochon, P.; Natansohn, A. J. Chem. Phys. 1998, 109, 1505. (6) (a) Kumar, J.; Li, L.; Jiang, X. L.; Kim, D. Y.; Lee, T. S.; Tripathy, S. K. Appl. Phys. Lett. 1998, 72, 2096. (b) Bian, S.; Williams, J. M.; Kim, D. Y.; Li, L.; Balasubramanian, S.; Kumar, J.; Tripathy, S. K. J. Appl. Phys. 1999, 86, 4498.

formation has been attributed to internal pressure gradients caused by an isomerization-driven free volume expansion in the bulk,5 the force based on the dipolar interaction of the azo chromophores with the optically induced electric field gradient,6 a translational wormlike diffusion caused by photoisomerization of the azobenzene chromophores,7 a mean-field force caused by molecule alignment among others.8 Recent studies show that a photomechanical effect occurs in thin films of azo polymers, which appears to be a new candidate mechanism to explain formation of SRGs on azo polymer films.9 Although the exact mechanism of SRG formation is still a debatable issue, it is generally agreed that the reversible surface deformation is caused by mass migration of azo polymers upon light irradiation. The discovery implies that the light force could drive the azo polymers to move at least within the submicrometer scale. Recently, the photoinduced deformation, which could be attributed to photodriven mass migration, has been observed for colloidal spheres and microwires made of azo polymers.10,11 Upon irradiation of the polarized Ar+ laser single beam or interfering beams, the objects can be stretched along the polarization direction of the light. As a logical expectation, this effect should also be observed for various micrometer-scale objects composed of azo polymers. Exploration along this line can supply new insight into the photoinduced effect, which can result in new approaches for obtaining various photoprocessible nano/micrometer-scale objects. (7) Lefin, P.; Fiorini, C.; Nunzi, J. M. Pure Appl. Opt. 1998, 7, 71. (8) Pedersen, T. G.; Johansen, P. M.; Holme, N. C. R.; Ramanujam, P. S.; Hilvsted, S. Phys. ReV. Lett. 1998, 80, 89. (9) (a) Tanchak, O. M.; Barrett, C. J. Macromolecules 2005, 38, 10566. (b) Yager, K. G.; Tanchak, O. M.; Godbout, C.; Fritzsche, H.; Barrett, C. J. Macromolecules 2006, 39, 9311. (c) Yager, K. G.; Barrett, C. J. Macromolecules 2006, 39, 9320. (10) (a) Li, Y. B.; He, Y. N.; Tong, X. L.; Wang, X. G. J. Am. Chem. Soc. 2005, 127, 2402. (b) Li, Y. B.; He, Y. N.; Tong, X. L.; Wang, X. G. Langmuir 2006, 22, 2288. (11) Liu, B.; He, Y. N.; Wang, X. G. Langmuir 2006, 22, 10233.

10.1021/la7016402 CCC: $37.00 © 2007 American Chemical Society Published on Web 09/20/2007

Azo Polymer Microspherical Cap Array

Soft-lithographic methods can be used to prepare different nano/micrometer-scale surface patterns and structures.12-15 The patterns and structures are fabricated using soft processing tools such as elastomeric stamps, molds, and masks. The tools can be prepared by the replica molding of poly(dimethylsiloxane) (PDMS) against masters that have patterned structures on the surfaces. An array of microspherical caps or microhemispheres is one type of surface relief structures with special importance, which can be used as microlens arrays to mimic the function of the compound eyes of insects.16-18 Due to its unique properties, microlens shows tremendous application potential in organic light-emitting diodes,19 charge-coupled device cameras,20 and other microoptical systems.21 Recently, it has been reported that microhemisphere arrays can be prepared by the soft-lithographic method.22-24 In one of the methods, PDMS elastomer with hemispherical air void arrays was prepared and the microlens array was obtained by imprinting on ultraviolet-curable photopolymers and then curing with UV light.22 Arrays of microspherical caps or microhemispheres made of azo polymers can potentially be used as a novel-type microlens system with photoprocessible properties. Moreover, the arrays can be ideal media for studying the photoinduced deformation behavior of azo polymers because of the well-defined shape and alignment. To our knowledge, no report concerning the fabrication and photoresponsive propertiesof such arrays has appeared in the literature. In this work, azo polymer microspherical cap arrays were fabricated by a soft-lithographic contact printing method. Elastomeric stamps were prepared by casting PDMS precursor against hexagonal close-packed arrays of polystyrene (PS) colloidal spheres. The “ink” was a THF solution of an epoxybased azo polymer (BP-AZ-CA) with a proper concentration. The microspherical cap arrays were fabricated by printing with the “inked” stamps on substrates. The spherical cap size and interval can be adjusted by altering the diameters of PS spheres and printing conditions. Upon irradiation of a polarized Ar+ (488 nm) laser single beam, the microspherical caps could be stretched along the polarization direction of the light. When irradiated with interfering Ar+ laser beams, more complicated surface relief structures could be inscribed. The details concerning the fabrication and photoresponsive properties will be reported herein. (12) (a) Xia, Y.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1998, 37, 550. (b) Zhao, X.; Xia, Y.; Whitesides, G. M. J. Mater. Chem. 1997, 7, 1069. (13) Gates, B. D.; Xu, Q.; Stewart, M.; Ryan, D.; Willson, C. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1171. (14) Xia, Y.; Kim, E.; Zhao, X. M.; Rogers, J. A.; Prentiss, M.; Whitesides, G. M. Science 1996, 273, 347. (15) (a) Bowden, N.; Huck, W. T. S.; Paul, K. E.; Whitesides, G. M. Appl. Phys. Lett. 1999, 75, 2557. (b) Quake, S. R.; Scherer, A. Science 2000, 290, 1536. (c) Wilbur, J. L.; Jackman, R. J.; Whitesides, G. M.; Cheung, E. L.; Lee, L. K.; Prentiss, M. G. Chem. Mater. 1996, 8, 1380. (d) Rogers, J. A.; Jackman, R. J.; Schueller, O. J. A.; Whitesides, G. M. Appl. Opt. 1996, 35, 6641. (16) Lee, L. P.; Szema, R. Science 2005, 310, 1148. (17) Hutley, M. C. J. Opt. A: Pure Appl. Opt. 1999, 1, 790. (18) Land, M. F. Pure Appl. Opt. 1997, 6, 599. (19) (a) Moller, S.; Forrest, S. R. J. Appl. Phys. 2002, 91, 3324. (b) Sun, Y.; Forrest, S. R. J. Appl. Phys. 2006, 100, 073106. (20) Ke, C. J.; Yi, X. J.; Lai, J. J.; Chen, S. H.; He, M. Int. J. Infrared Millim. WaVes 2004, 25, 439. (21) (a) Roulet, J. C.; Volkel, R.; Herzig, H. P.; Verpoorte, E.; Rooij, N. F.; Dandliker, R. Opt. Eng. 2001, 40, 814. (b) Kim, J.; Nayak, S.; Lyon, L. A. J. Am. Chem. Soc. 2005, 127, 9588. (c) Wu, M-H.; Paul, K. E.; Yang, J.; Whitesides, G. M. Appl. Phys. Lett. 2002, 80, 3500. (22) Nam, H. J.; Jung, D. Y.; Yi, G. R.; Choi, H. Langmuir 2006, 22, 7358. (23) Xia, Y.; Tien, J.; Qin, D.; Whitesides, G. M. Langmuir 1996, 12, 4033. (24) Campbell, C. J.; Smoukov, S. K.; Bishop, K. J. M.; Baker, E.; Grzybowski, B. A. AdV. Mater. 2006, 18, 2004.

Langmuir, Vol. 23, No. 22, 2007 11267

Experimental Section Materials. The chemical structure of the epoxy-based azo polymer (BP-AZ-CA) is given as follows, which contains the pseudo-stilbenetype azo chromophore at each repeating unit. The synthesis and

characterization of the polymer have been reported elsewhere.25 The BP-AZ-CA sample used in this study possessed a numberaverage molecular weight of 41 000 with a polydispersity index of 2.2. The PDMS elastomer kit was bought from Dow Corning (Sylgard 184). Ammonium persulfate ((NH4)2S2O8), obtained from the commercial source, was recrystallized twice with deionized water before use. The glass slides were ultrasonically treated in deionized water for 10 min, rinsed with acetone, and blown dry. Other chemicals and solvents were commercially purchased and used without further purification. Characterization. The sizes and size distributions of the colloidal spheres were measured with a Malvern Zetasizer 3000 dynamic light scattering instrument. The atomic force microscopy (AFM) images were obtained using a scanning probe microscope (Nanoscope IIIa, Digital Instrument) in tapping mode. The scanning electron microscopy (SEM) images were obtained by applying a Hitachi S-4500 electron microscope operated at an accelerating voltage of 15 kV. The samples prepared for SEM studies were coated with a thin layer of gold (∼10 nm in thickness) before measurement. Transmission electron microscopy (TEM) images were recorded on a JEM-1200EX (JEOL) microscope at an accelerating voltage of 100 kV. The TEM samples were prepared by transferring the floated microspherical cap thin films (obtained after HF etching) onto the copper grids. Preparation of Polystyrene (PS) Colloids. The PS colloidal spheres were synthesized according to the literature.26 Styrene (9.6 g) and methacrylic acid (0.4 g) were mixed in a round-bottom flask, and NaHCO3 (0.24 g) dissolved in 80 mL of deionized water was added into the flask. The mixture was mechanically stirred for about 20 min and gradually heated to 60 °C in a water bath. Then (NH4)2S2O8 (0.08 g) in 20 mL of water was added as the initiator. The system was stirred and reacted at 80 °C for 10 h. The resulting latex was separated by filtration. The average hydrodynamic diameter and polydispersity index of the spheres measured by DLS were 344 nm and 0.009, respectively. The average diameter estimated from TEM was 300 nm (average of 50 particles). The spheres with sizes of 175 and 440 nm (measured by DLS) were also synthesized through the same procedure by changing the feed ratio of styrene to methacrylic acid to 9:1 and 9.7:0.3 (wt:wt), respectively. Master Fabrication. Hexagonal close-packed 2D colloidal arrays of PS spheres were used as masters for PDMS replication. The colloidal arrays were prepared through vertical deposition.27 In the process, clean glass slides were immersed vertically into beakers containing water suspensions of the PS spheres (about 5 mg/mL). The liquid surface was controlled to decline slowly (about 0.35 µm/s) by removing water with a stepmotor-driven syringe at room temperature. Under the attraction of capillary force at the meniscus, hexagonal close-packed 2D arrays were formed on the glass substrates. PDMS Stamp Preparation. The elastomeric stamps were prepared by replica molding. Poly(dimethylsiloxane) (PDMS) precursor polymer was obtained by mixing the elastomer base and curing agent (Sylgard 184, Dow Corning) in a proper ratio (10:1, (25) He, Y. N.; Wang, X. G.; Zhou, Q. X. Polymer 2002, 43, 7325. (26) Furusawa, K.; Norde, W.; Lyklema, J. Kolloid Z. Z. Polym. 1972, 250, 908. (27) (a) Jiang, P.; Bertone, J. F.; Hwang, K. S.; Colvin, V. L. Chem. Mater. 1999, 11, 2132. (b) Goldenberg, L. M.; Wagner, J.; Stumpe, J.; Paulke, B. R.; Gornitz, E. Langmuir 2002, 18, 3319.

11268 Langmuir, Vol. 23, No. 22, 2007

Liu et al.

Figure 1. Schematic representation of the preparation of photoprocessible microspherical cap arrays and some control parameters: (1) casting PDMS precursor against the PS colloidal array obtained from vertical deposition, (2) obtaining the PDMS stamp after curing and peeling off, (3) printing with the “inked” stamp on the substrate, (4) formation of the microspherical cap array after solidification and peeling off the stamp, (5) irradiation with polarized Ar+ laser single beam, (6) irradiation with interfering p-polarized Ar+ laser beams. wt/wt). The precursor was cast against the masters, which required about 30 min for the precursor to effectively penetrate into the voids of PS arrays. Then, the samples were cured in a 60 °C oven for 4 h. After curing the PDMS stamps were peeled off and used for the contact printing application. Microspherical Cap Array Fabrication. The stamps were “inked” with a THF solution of BP-AZ-CA (concentration in the range from 4 to 20 mg/mL), and excess liquid on the surfaces was removed by carefully wiping. The PDMS stamps were printed on clean glass slides, pressing the “inked” surfaces conformally against the substrates for a few seconds (ca. 10 s). After that the stamps were peeled off, and the substrates covered with the printed patterns were kept in a 40 °C oven for 2 h. Laser Irradiation. An Ar+ laser beam (488 nm) was spatially filtered, expanded, and collimated to produce the beam with uniform intensity (80 mW/cm2). For the single-beam irradiation experiment, the linearly polarized laser beam was incident perpendicularly to the substrate surfaces containing the microspherical cap arrays. To carry out the interfering beam irradiation experiments, the microspherical cap arrays were irradiated by the p-polarized interfering Ar+ laser beams. The p-polarized interfering laser beams were obtained by splitting the single beam into two halves: one-half of the beam was incident on the arrays directly, and the other half of the beam was reflected onto the arrays from a mirror. This experimental setup was similar to that reported for SRG fabrication.3,25 The experiments were carried out under ambient conditions.

Results and Discussion The microspherical cap arrays were prepared by softlithographic contact printing. The major steps to fabricate the structures and some controlled parameters are illustrated in Figure 1. The hexagonal close-packed 2D array of PS colloid spheres, prepared through vertical deposition, is used as the master. PDMS precursor is cast against the master and cured properly (Step 1). The upper hemispherical feature on the master is complementarily transferred to the PDMS surface. After peeling off, the PDMS stamp featured with hemispherical air void array on the surface is obtained (Step 2). The PDMS stamp is “inked” with a THF solution of BP-AZ-CA and pressed against a glass slide for a proper time period (Step 3). After peeling off the stamp and drying in an oven, the microspherical cap array is constructed on the glass slide (Step 4). The spherical cap array can be manipulated with light irradiation. To study the photoinduced shape deformation, the printed spherical cap array is irradiated with a linearly polarized Ar+ laser single beam (Step 5) or

Figure 2. AFM and SEM images of the colloidal masters and PDMS stamps: (a) AFM image of PS colloidal array, Φ0 ) 299 nm, d0 ) 316 nm, (b) SEM image of PS colloidal array, Φ0 ) 310 nm, d0 ) 308 nm, (c) AFM image of the PDMS stamp surface, Φ1 ) 240 nm, d1 ) 310 nm, (d) SEM image of the PDMS stamp surface, Φ1 ) 252 nm, d1 ) 307 nm.

Figure 3. AFM and SEM images of the BP-AZ-CA microspherical cap arrays: (a) AFM image, the “ink” concentration was 4 mg/mL, Φ2 ) 231 nm, d2 ) 322 nm, (b) SEM image, the “ink” concentration was 4 mg/mL, Φ2 ) 220 nm, d2 ) 310 nm, (c) AFM image, the “ink” concentration was 20 mg/mL, Φ2 ) 258 nm, d2 ) 306 nm, (d) SEM image, the “ink” concentration was 20 mg/mL, Φ2 ) 264 nm, d2 ) 308 nm.

interfering p-polarized Ar+ laser beams (Step 6). The effects of light irradiation on the microspherical caps are illustrated in the same figure. The fabrication and results obtained from the light irradiation study will be presented in the following parts in detail. Microspherical Cap Array Fabrication. The masters for casting PDMS stamps were 2D arrays of the PS colloidal spheres prepared by the vertical deposition method. In a typical case, PS colloidal spheres with an average diameter of 300 nm were used as the building blocks. The glass substrates were immersed vertically into the water suspension (5 mg/mL) of the PS spheres. When the liquid level decreased very slowly (about 0.35 µm/s),

Azo Polymer Microspherical Cap Array

Figure 4. SEM images of the BP-AZ-CA microspherical cap arrays after irradiation with polarized Ar+ laser single beam (488 nm, 80 mW/cm2): (a) for 1 h, d3 ) 314 nm, l1/l2 ) 1.23, (b) enlarged image of a, (c) for 2 h, d3 ) 311 nm, l1/l2 ) 1.39, (d) enlarged image of c, (e) for 3 h, d3 ) 311 nm, l1/l2 ) 1.53, (f) enlarged image of e. The spherical cap arrays were printed with the “ink” of 4 mg/mL.

the PS spheres were attracted and arranged at the meniscus under the effect of the capillary force, which formed hexagonal closepacked 2D arrays on the glass substrates. Figure 2a and b shows some typical AFM and SEM images of those well-aligned PS arrays. From the images, the average sphere diameters (Φ0) are measured to be 299 (AFM) and 310 nm (SEM); the average sphere center distances (d0) are 316 (AFM) and 308 nm (SEM). The Φ0 and d0 data are consistent with the average diameter of the spheres obtained by TEM. The PDMS stamps were obtained by cast liquid PDMS precursor against the featured surfaces of PS colloidal arrays. Figure 2c and d gives some typical AFM and SEM images of the PDMS stamp surfaces. The average void center distances (d1) are measured to be 310 (AFM) and 307 nm (SEM). This result demonstrates that the framework of the colloidal arrays is negatively replicated by the stamps. The top-view circle diameters (Φ1) of the voids are estimated to be 240 (AFM) and 252 nm (SEM), respectively, and the average depth of voids (h1) estimated from AFM is about 89 nm. In the replica molding the tiny rift between spheres inhibited penetration of the viscous PDMS precursor; therefore, only the upper hemispheres could be wetted and mold. As a result, the voids on the PDMS surface became shallower. In the soft-lithographic printing step, THF solutions of BPAZ-CA with different concentrations were used as “inks”. For comparison, two kinds of “inks” with concentrations of 4 and 20 mg/mL were applied. The PDMS stamps were “inked” by touching the polymer solution for about 1 s, which left a thin liquid layer adsorbed on the patterned surfaces. Although the “ink” concentration was low, the solvent of “inked” thin liquid layer vaporized very quickly and a highly concentrated and

Langmuir, Vol. 23, No. 22, 2007 11269

Figure 5. AFM images of the microspherical cap arrays after irradiation with polarized Ar+ laser single beam (488 nm, 80 mW/ cm2): (a) for 1 h, d3 ) 320 nm, l1/l2 ) 1.27, the “ink” concentration was 4 mg/mL; (b) for 2 h, d3 ) 318 nm, l1/l2 ) 1.42, the “ink” concentration was 4 mg/mL; (c) for 1 h, d3 ) 308 nm, l1/l2 ) 1.17, the “ink” concentration was 20 mg/mL; (d) for 2 h, d3 ) 308 nm, l1/l2 ) 1.20, the “ink” concentration was 20 mg/mL.

Figure 6. SEM images of the microspherical cap arrays after irradiation with polarized Ar+ laser single beam (488 nm, 80 mW/ cm2): (a) for 1 h, d3 ) 314 nm, l1/l2 ) 1.15; (b) for 2 h; (c) for 3 h; (d) for 40 min, d3 ) 310 nm, section view. The microspherical cap arrays were printed with the “ink” of 20 mg/mL.

viscous thin film formed on the stamp surfaces. By pressing the “inked” surfaces of the stamps against clean glass slides for a few seconds (ca. 10 s), the BP-AZ-CA material was transferred to the slide surfaces and shaped by the voids on the stamps. After peeling off the PDMS stamps, the samples were kept in a 40 °C oven for 2 h before subsequent experiments. Figure 3 shows some typical AFM and SEM images of the printed surfaces. Microspherical cap arrays in Figure 3a and b were printed with the “ink” of 4 mg/mL. The top-view circle diameters (Φ2) of the spherical caps estimated from the images are 231 (AFM) and 220 nm (SEM), and the average distances between the spherical caps (d2) are 322 (AFM) and 310 nm (SEM), respectively. The average height of the spherical caps

11270 Langmuir, Vol. 23, No. 22, 2007

Liu et al.

Figure 7. SEM (a-d) and AFM (e, f) images of the microspherical cap arrays after irradiation by p-polarized interfering Ar+ laser beams: (a) for 20 min, d4 ) 314 nm, p ) 805 nm, (b) enlarged image of a, (c) for 45 min, d4 ) 311 nm, p ) 816 nm, (d) for 1 h, p ) 773 nm, (e) for 45 min, d4 ) 310 nm, p ) 806 nm, (f) for 1 h, d4 ) 313 nm, p ) 761 nm. The microspherical cap arrays were printed with the “ink” of 20 mg/mL.

(h2) in Figure 3a is about 69 nm. Figure 3c and d gives the images of spherical cap arrays printed with the “ink” of 20 mg/mL. Φ2 is estimated to be 258 (AFM) and 264 nm (SEM), and d2 is 306 (AFM) and 308 nm (TEM), respectively. The average height of microspherical caps in Figure 3c is about 78 nm. TEM images of some microspherical cap arrays can be seen in Figure S1 (Supporting Information). Obviously, the variation in “ink” concentration resulted in some differences in the microspherical cap size and morphology. If the concentration of the “ink” is low, such as 4 mg/mL, the diameter and height of the microspherical caps are smaller (Figure 3a and b). Increasing the concentration to 20 mg/mL, uniform microspherical caps with a larger diameter and height can be obtained (Figure 3c and d). This difference can be attributed to shrinkage of the printed pattern after removing the remaining solvent. When the “ink” concentration varied in the range from 10 to 30 mg/mL, the resulting microspherical caps showed almost the same size and surface morphology. When the concentration was further increased, transferring the “inked” layer through the contact printing became more and more difficult. Response to Single Laser Beam Irradiation. One interesting property of BP-AZ-CA is its photoresponsive nature related to

the azo chromophores. The microspherical cap arrays can be used as an ideal medium to study the photoinduced deformation effect. In the single-beam irradiation experiment the spherical cap arrays were irradiated by a linearly polarized Ar+ laser beam (488 nm, 80 mW/cm2) incident vertically upon the sample for different time periods. Figure 4 gives some typical SEM images of the microspherical cap arrays after irradiation for 1 h (Figure 4a,b), 2 h (Figure 4c,d), and 3 h (Figure 4e,f). The arrays were printed using the “ink” with a concentration of 4 mg/mL. The corresponding AFM images of the same type samples after irradiating for 1 and 2 h are given in Figure 5a and b. The images show that the spherical caps are elongated along one direction. By comparing with the polarization direction of the light it can be concluded that the spherical caps are elongated along the polarization direction. The deformation degree can be estimated from the SEM images. For the spherical caps after irradiating for 1 h, the average major axis (l1) and minor axis (l2) of the top-view ellipses are estimated to be 265 and 215 nm, and the axial ratio (l1/l2) is 1.23. After irradiation for 2 h, the l1, l2, and axial ratio (l1/l2) are 290 nm, 208 nm, and 1.39. After irradiation for 3 h, the axial ratio further increases to 1.53. From the SEM images, the average distances between centers of the deformed

Azo Polymer Microspherical Cap Array

spherical caps (d3) are estimated to be 314, 311, and 311 nm after irradiation for 1, 2, and 3 h. From the AFM images, the l1, l2, and l1/l2 of the top-view ellipses are 271 nm, 214 nm, and 1.27 for 1 h irradiation and 291 nm, 205 nm, and 1.42 for 2 h irradiation. The d3 values are estimated to be 320 and 318 nm for 1 and 2 h irradiation. These data are close to those obtained from SEM measurement. The heights of spherical caps (h3) are 56 and 48 nm after 1 and 2 h irradiation. The above results show that the spherical caps are significantly elongated in the polarization direction and the heights of the spherical caps decrease but the average distances (d3) between the microspherical cap centers remain the same. Figures 5c,d and 6 show the AFM and SEM images of the microspherical caps prepared using the “ink” with a concentration of 20 mg/mL after irradiating with the same light source for different time periods. The spherical caps show similar deformation along the polarization direction of the light. However, due to the narrower space between the spherical caps, deformation is restricted by the neighboring spherical caps. After irradiation for 1 h (Figures 5c and 6a), l1, l2, and l1/l2 of the top-view ellipses are estimated to be 280 nm, 237 nm, and 1.17 from AFM and 291 nm, 254 nm, and 1.15 from SEM. d3 and h3 are estimated to be 308 and 52 nm from AFM, and d3 is 314 nm as measured from SEM. Figures 5d and 6b show that after 2 h irradiation the deformed microspherical caps start to merge. After irradiation for 3 h (Figure 6c) the deformed microspherical caps almost completely merge together, which look like the thin film under SEM observation. Figure 6d shows the SEM section-view of the microspherical caps after Ar+ laser single-beam irradiation for 40 min. l1, d3, and h3 are estimated to be 270, 310, and 63 nm. Microspherical cap arrays with different sizes, fabricated using colloidal spheres of 175 and 440 nm (measured by DLS) as the masters, could also be stretched under irradiation of polarized Ar+ laser single beam (Figure S2, Supporting Information). Response to Interfering Beam Irradiation. The microspherical cap arrays can show alternately varied modulations after exposure to the interfering patterns of p-polarized Ar+ laser beams. The interference pattern of Ar+ laser beams was produced by a similar experimental setup used for writing SRGs.3,25 The grating period of the interference pattern was adjusted by changing the intersection angle between the interfering beams. In this study the grating period of the interference pattern was adjusted to be around 800 nm and the intensity of the laser beam was 80 mW/ cm2. After exposure to interfering p-polarized Ar+ laser beams for a period of time, the samples were characterized using SEM and AFM. Figure 7 gives the SEM and AFM images of the microspherical cap arrays after interfering beam irradiation. The arrays were printed using the BP-AZ-CA “ink” with a concentration of 20 mg/mL. After irradiation for 20 min, SEM observation shows that some microspherical caps are elongated along the direction parallel to the grating vector and the heights of the spherical caps decrease at the same time (Figure 7a and b). As the polarization direction coincides with the grating vector, elongation is parallel to the polarization direction. Comparing with the previous study on SRG formation,6,25 it can be concluded that the deformed areas correspond to the bright regions of the interfering fringes. The average distance between the spherical cap centers (d4) and the distance between the undeformed spherical cap lines (p) are estimated to be 314 and 805 nm. After irradiation for 45 min, more significant deformation can be observed for the microspherical caps exposed to the light field (Figure 7c). d4 and p are estimated to be 311 and 816 nm from the SEM images. After irradiation for 1 h, only undeformed or slightly deformed

Langmuir, Vol. 23, No. 22, 2007 11271

Figure 8. SEM images of the microspherical cap arrays after irradiation by p-polarized interfering Ar+ laser beams: (a) for 1 h, d4 ) 313 nm, p ) 774 nm; (b) an enlarged image of a. The microspherical cap arrays were printed with the “ink” of 4 mg/mL.

microspherical caps can be observed in SEM images due to loss of topographic contrast of the deformed spherical caps (Figure 7d). p is estimated to be 773 nm from the SEM images. AFM observation further indicates that the heights of undeformed microspherical caps are larger than those of deformed ones (Figure 7e,f). The microspherical caps in the deformed regions are stretched along the polarization direction. After irradiation for 45 min, d4 and p are estimated to be 310 and 806 nm from the AFM images. After irradiation for 1 h, d4 and p are estimated to be 313 and 761 nm from the AFM images. In all the above cases the average distances between the microspherical caps almost remain unchanged after light irradiation. The distances between the undeformed spherical cap lines are consistent with the interfering fringe periods. For microspherical caps obtained using the “ink” with a concentration of 4 mg/mL a similar photoinduced deformation behavior can be observed upon exposure to the interfering pattern. Figure 8 shows some typical SEM images of the deformed microspherical cap arrays. Because of the larger space between the microspherical caps in this case, the deformed spherical caps can be more clearly viewed. After light irradiation for 1 h, the microspherical caps exposed to the bright regions of the interference pattern change to ellipsoidal caps. d4 and p are estimated to be 313 and 774 nm, respectively. Like SRG formation, the mechanism for photoinduced deformation of microspherical caps is not fully understood at the current stage. The above study indicates that light irradiation can cause similar stretching effects to those observed for azo polymer colloidal spheres, although the deformation rate is obviously slower for the microspherical caps.10 As the microspherical caps are tightly adhered on the substrates, the slower deformation rate could be attributed to surface adhesion and the difference of geometrical shapes in both cases. A clear distinction between these two factors cannot be drawn from the above studies. Since the distances between the microspherical caps are not altered by the interfering beam irradiation, deformation should behave more like viscous flow on the surfaces. This can also be supported by the height decrease during deformation. This deformation nature is obviously different from the optomechanical deformation observed for free-standing liquid-crystal elastomeric thin films,

11272 Langmuir, Vol. 23, No. 22, 2007

which can respond in a much shorter time period.4d,e In the interfering laser beam irradiation process, flowing can be caused by the optically induced electric field gradient related to the light intensity variation crossing the interference fringes.6 For the case of irradiation with the linearly polarized laser single beam, deformation is more difficult to explain and is believed to be related to the geometric shape of microspherical caps. In both cases, the repeated trans-cis-trans isomerization cycles of the azo chromophores play a key role to soften the materials through free volume expansion.5 The method reported above can be used to prepare various surface patterns using microspherical caps with different sizes and carefully selecting light irradiation parameters. Also, this method can potentially be used as a new approach to fabricate microlens arrays with photoprocessible nature. As one of the advantages, the curvature and shape of the lens can be adjusted by light irradiation under some carefully controlled conditions.

Conclusion In summary, azo polymer microspherical cap arrays have been fabricated through soft-lithographic contact printing using

Liu et al.

colloidal arrays as masters and duplicated PDMS elastomer as stamps. The size of the microspherical caps can be feasibly altered by adjusting the size of PS colloids and the “ink” concentration. The spherical cap arrays show unique photoprocessible nature upon Ar+ laser irradiation. After irradiation of the linearly polarized Ar+ laser single beam, the microspherical caps can be elongated along the polarization direction. Upon irradiation with interfering Ar+ laser beams, only the spherical caps in the bright regions of the interference patterns are deformed. The method developed in this work can potentially be used as a new approach for fabricating photoprocessible microlens arrays and various other surface patterns. Acknowledgment. The financial support from NSFC under Projects 50533040 and 20374033 is gratefully acknowledged. Supporting Information Available: More results obtained from the structure characterization. This material is available free of charge via the Internet at http://pubs.acs.org. LA7016402