Polymer-on-Polymer Stamping - American Chemical Society

Mar 20, 2002 - Xueping Jiang, Haipeng Zheng,† Shoshana Gourdin,† and Paula T. Hammond*. Department of Chemical Engineering, Massachusetts Institut...
5 downloads 0 Views 751KB Size
Langmuir 2002, 18, 2607-2615

2607

Polymer-on-Polymer Stamping: Universal Approaches to Chemically Patterned Surfaces Xueping Jiang, Haipeng Zheng,† Shoshana Gourdin,† and Paula T. Hammond* Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Received July 16, 2001. In Final Form: January 10, 2002 A new approach to create chemically patterned surfaces utilizing polymers and copolymers is introduced. In this approach, chemical patterns are achieved by the direct stamping of functional polymers onto a surface containing complementary functional groups. The resulting pattern is then used as a template for the further deposition of materials on the surface. This concept can be applied to various functional polymer and substrate systems as well as different thin film deposition techniques. This approach is demonstrated with the direct stamping of polystyrene-poly(acrylic acid) block copolymers (PS-PAA) to create alternating hydrophobic/hydrophilic regions and polyelectrolytes to create alternating positively and negatively charged regions. This approach has been used for patterning surfaces and templating materials deposition. When a patterned polyelectrolyte film is used as the base layer or substrate in this process, functionality can be incorporated in the underlying layer, making this approach particularly relevant to device and sensor applications. This approach is universal to a number of substrates, many of which can be used to adsorb a polyelectrolyte layer to provide a functional surface. Various substrates such as Si, glass, and plastic can be patterned with this method with relative ease, and without the need for traditional alkanethiol or silane monolayers. Factors such as stamping temperature, contact time, and substrate pretreatment on the nature of the transferred pattern have been investigated and will be discussed.

Introduction The use of micrometer scale printing techniques to form chemical templates was first introduced by the Whitesides group using the transfer of alkanethiols1-3 onto gold surfaces with a poly(dimethylsiloxane) elastomeric stamp. Microcontact printing of self-assembled monolayers (SAMs) was extended to the printing of trichlorosilanes,4 which form monolayers on silicon and other metal oxide surfaces. The use of the microcontact printing method has revolutionized the area of materials science by presenting a fast, easy approach to chemical patterning. The technique has been used to create masks for the etching of silicon, gold, and other metals5,6 to form polymer microarrays7 and template electrochemical deposition.8 By patterning different functional systems on the surface, this approach has been used to template living polymerization from surfaces,9-13 catalysts for electroless deposition,14 liquid * To whom correspondence should be addressed: fax, 617-2585766; phone, 617-258-7577; e-mail, [email protected]. † These authors contributed equally to this work. (1) Kumar, A.; Whitesides, G. M. Science 1994, 263, 60-62. (2) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498-1511. (3) Xia, Y.; Whitesides, G. M. J. Am. Chem. Soc. 1995, 117, 32743275. (4) Xia, Y.; Mrksich, M.; Kim, E.; Whitesides, G. M. J. Am. Chem. Soc. 1995, 117, 9576-9577. (5) Whidden, T. K.; Ferry, D. K.; Kozicki, M. N.; Kim, E.; Kkumar, A.; Wilbur, J. L.; Whitesides, G. M. Nanotechnology 1996, 7, 447-451. (6) Kim, E.; Kumar, A.; Whitesides, G. M. J. Electrochem. Soc. 1995, 142, 628-633. (7) Kim, E.; Whitesides, G. M.; Lee, L. K.; Smith, S. P.; Prentiss, M. Adv. Mater. 1996, 8, 139. (8) Gorman, C. B.; Biebuyck, H. A.; Whitesides, G. M. Chem. Mater. 1995, 7, 526-529. (9) Husemann, M.; Morrison, M.; Benoit, D.; Frommer, J.; Mate, C. M.; Hinsberg, W. D.; Hedrick, J. L.; Hawker, C. J. J. Am. Chem. Soc. 2000, 122, 1844-1845. (10) Shah, R. R.; Merreceyes, D.; Husemann, M.; Rees, I.; Abbott, N. L.; Hawker, C. J.; Hedrick, J. L. Macromolecules 2000, 33, 597-605. (11) Husemann, M.; Mecerreyes, D.; Hawker, C. J.; Hedrick, J. L.; Shah, R.; Abbott, N. L. Angew. Chem., Int. Ed. 1999, 38, 647-649.

crystals,15 and even cell growth on patterned surfaces.16-22 In past work, our group has presented the ability to pattern polymeric thin films in situ through the use of chemically patterned surfaces as templates for ionic multilayer assembly.23-27 Among the most important aspects of microcontact printing is the fact that it is nonlithographic, eliminating the cost of expensive photolithographic processes; further, the process is easily accessible and can be practiced under ordinary lab benchtop conditions. Ultimately, the ability to pattern chemistry provides a route (12) Kim, N. Y.; Jeon, N. L.; Choi, I. S.; Takami, S.; Harada, Y.; Finnie, K. R.; Girolami, G. S.; Nuzzo, R. G.; Whitesides, G. M.; Laibinis, P. E. Macromolecules 2000, 33, 2793-2795. (13) Jeon, N. L.; Choi, I. S.; Whitesides, G. M.; Kim, N. Y.; Laibinis, P. E.; Harada, Y.; Finnie, K. R.; Girolami, G. S.; Nuzzo, R. G. Appl. Phys. Lett. 1999, 75, 4201-4203. (14) Hidber, P. C.; Nealey, P. F.; Helbig, W.; Whitesides, G. M. Langmuir 1996, 12, 5209-5215. (15) Gupta, V. K.; Abbott, N. L. Science 1997, 276, 1533-1536. (16) Mrksich, M.; Chen, C. S.; Xia, Y. N.; Dike, L. E.; Ingber, D. E.; Whitesides, G. M. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 10775-10778. (17) Mrksich, M.; Dike, L. E.; Tien, J.; Ingber, D. E.; Whitesides, G. M. Exp. Cell Res. 1997, 235, 305-313. (18) Zhang, S.; Yan, L.; Altman, M.; Lassle, M.; Nugent, H.; Frankel, F.; Lauffenburger, D. A.; Whitesides, G. M.; Rich, A. Biomaterials 1999, 20, 1213-1220. (19) Thomas, K. J.; Crooks, R. M. Abs. Paper - Am. Chem. Soc. 221st 2001, COLL-499. (20) Amirpour, M. L.; Ghosh, P.; Lackowski, W. M.; Crooks, R. M.; Pishko, M. V. Anal. Chem. 2001, 73, 1560-1566. (21) Craighead, H. G.; Turner, S. W.; Davis, R. C.; James, C.; Perez, A. M.; St. John, P. M.; Isaacson, M. S.; Kam, L. S., W.; Turner, J. N.; Banker, G. Biomed. Microdevices 1998, 1, 49-64. (22) Lu, L.; Kam, L.; Hasenbein, M.; Nyalakonda, K.; Bizios, R.; Gopferich, A.; Young, J. F.; Mikos, A. G. Biomaterials 1999, 20, 23512361. (23) Hammond, P. T.; Whitesides, G. M. Macromolecules 1995, 28, 7569-7571. (24) Clark, S. L.; Montague, M.; Hammond, P. T. Supramol. Sci. 1997, 4, 141-146. (25) Clark, S. L.; Montague, M. F.; Hammond, P. T. Macromolecules 1997, 30, 7237-7244. (26) Clark, S. L.; Hammond, P. T. Adv. Mater. 1998, 10, 1515-1519. (27) Clark, S. L.; Handy, E. S.; Rubner, M. F.; Hammond, P. T. Adv. Mater. 1999, 11, 1031-1035.

10.1021/la011098d CCC: $22.00 © 2002 American Chemical Society Published on Web 03/20/2002

2608

Langmuir, Vol. 18, No. 7, 2002

Figure 1. Schematic illustrating the transfer of a functional polymer to a surface with complementary functionality using polymer-on-polymer stamping.

to surface-directed assembly, in which surface functional groups can be used to direct etching, deposition, reaction, or other materials processes. Thus far, much of microcontact printing has been directed toward metal, glass, or semiconductor surfaces. More recently, other molecular systems such as polymers and ligands have been stamped onto surfaces; in these cases, the molecules were stamped onto a reactive alkanethiolate SAM.28,29 Similar approaches have also been used to stamp proteins onto a chemically activated polymer surface.30 A number of interesting and important materials systems consist of plastic and other polymeric systems. It would be advantageous to be able to pattern functional chemistry onto a broader range of surfaces, including soft materials, without the use of high energy or other methods to pretreat the polymer surface. The concept of polymer-on-polymer stamping was first illustrated by stamping a functional graft polymer onto a polyelectrolyte multilayer film,31 demonstrating that graft copolymers can be directly patterned onto polyelectrolyte multilayer surfaces using the PDMS stamps commonly applied for microcontact printing. This first approach of patterning a polymer onto another polymeric system utilized polymers with anhydride groups that can react with polyamine surfaces. In this paper, we further broaden the use of polymer-on-polymer stamping to transfer surface chemistry to different regions of a substrate using block copolymers and common polyelectrolytes. By creating patterned functional chemistry atop a polyelectrolyte surface, we ultimately introduce a broad approach that enables modification of any surface which can be covered with at least one surface layer of polyion. Because a number of existing polymers may be used for this technique, many new functional surfaces can be achieved through the use of block copolymers or graft copolymers. The direct transfer of polymers using polymer-onpolymer stamping is illustrated schematically in Figure 1. A polymer is transferred directly to the top surface of a polyelectrolyte multilayer thin film based on covalent and/or noncovalent interactions. For example, a polyanion can be stamped directly onto a charged polycation top multilayer surface to create alternating regions of plus and minus charge. On the other hand, block and graft copolymers can present both a reactive functional group (28) Goetting, L. B.; Deng, T.; Whitesides, G. M. Langmuir 1999, 15, 1182-1191. (29) Lahiri, J.; Ostuni, E.; Whitesides, G. M. Langmuir 1999, 15, 2055-2060. (30) Yang, Z.; Chilkoti, A. Adv. Mater. 2000, 12, 413-417. (31) Jiang, X.-P.; Hammond, P. T. Langmuir 2000, 20, 8501-8509.

Jiang et al.

for attachment to the surface and a second functional group to modify the surface, as will be discussed further below. In either case, the goal is to create a stable functional monolayer thin film on the surface. The substrate may be glass, metal, or polymer/plastic, with a top functional surface which is either charged or contains reactive functional groups. In this work, polyelectrolyte multilayers are used to create functional surfaces which act as platforms for polymer stamping, as multilayers may be adsorbed on a number of different surfaces. Polyelectrolyte multilayers provide ideal surfaces because their surface functionality and charge density can be tailored using adsorption conditions. An added advantage of multilayer films is that additional functionality, including conducting or electroluminescent properties, can be incorporated in the film. Single polyelectrolyte layers adsorbed onto surfaces and charged oxide surfaces may also be used, as will be demonstrated here as well. Examination of the effectiveness and stability of polymeron-polymer stamping requires an understanding of the interactions between the polymer to be transferred and the underlying surface. Conditions must be achieved which allow for effective contact of the desired functional group with the chemical functional group on the surface, and both the polymer “ink” and surface functional groups must be in a form which encourages covalent, ionic, or hydrogen bonding. Aspects of polymer-on-polymer stamping will be discussed here using SAM-functionalized model surfaces and strong polyelectrolyte multilayer platforms as substrates. The influence of factors such as temperature and pH, as well as the stability of polymer-on-polymer stamping on model and multilayer surfaces will be addressed in this paper. A second important aspect of this technique is the optimization of the underlying polymer surface to achieve a high density of desired functional groups. The use of weak polyelectrolyte multilayers for this purpose provides many advantages, in that degree of segregation or interpenetration, as well as the density of surface functional groups such as acids or amines can be finely tuned through the variation of pH during the adsorption process.32 These aspects are further explored separately in ongoing collaborative investigations,33 in which weak polyelectrolyte multilayers are used to optimize the surface on which the polymer is transferred, as has been demonstrated with block copolymer adsorption onto multilayer surfaces.34 Experimental Section 1. Materials. Poly(diallyldimethylammonium chloride) (PDAC) of molecular weight MW ) 100 000-200 000, sulfonated polystyrene (SPS) of MW ) 70 000, poly(allylamine hydrochloride) (PAH) of MW ) 50000-65000, and poly(acrylic acid) (PAA) with MW ) 90 000 were purchased from Aldrich; polystyrene-poly(acrylic acid) diblock copolymer (PS-b-PAA) with a PS block MW ) 66 500 and PAA block MW ) 4500 was obtained from Polymer Source, Inc. The aminopropyltrimethoxysilane was also obtained from Aldrich. Poly(dimethylsiloxane) (PDMS) stamps were made by pouring a commercial two-component curable siloxane (Sylguard, 184 silicone elastomer kit) over a silicon master with the desired photoresist pattern.2 2. Stamping of PS-b-PAA Block Copolymer. Substrate Preparation. Three different substrates were used as platforms for the stamping of the PS-b-PAA block copolymer. Direct (32) Shiratori, S.; Rubner, M. F. Macromolecules 2000, 33, 42134219. (33) The use of weak polyelectrolyte multilayer platform is collaboration work between the Hammond and the Rubner research groups and will be reported separately. (34) Choi, J.; Rubner, M. F. J. Macromol. Sci.,- Pure Appl. Chem. 2001, A38, 1191-1206

Polymer-on-Polymer Stamping

Langmuir, Vol. 18, No. 7, 2002 2609 before inking. The polymer solution, or ink, was then applied to the stamp surface using a cotton swab that was wet with the ink. This thin layer of ink was then dried in air, or with N2 flow, and the stamp was placed on the multilayer platform and allowed to sit for a specified amount of time. Aqueous solutions of 20 mM PDAC and 0.1 M NaCl in water were used to stamp the polymer from aqueous solution. In this case, the stamping contact times ranged from 30 to 120 min. Ethanol/water mixtures were also used as inks. Five solvents of this type were tried: pure water, 75% water, 50% water, 25% water, and pure ethanol. The PDAC inks made with these solvents had concentrations of 0.025, 0.1, or 0.25 M (based on repeat unit). In this study, the stamping times were varied systematically from a few seconds to an hour for each ethanol/water combination. Following the stamping process, the patterned surface was rinsed thoroughly with deionized (DI) water applied directly to the film surface from a solvent squeeze bottle to remove any excess unbound polyelectrolyte. Characterization. A dye was used to visualize the stamped polyelectrolyte monolayer following the stamping and rinsing processes. The dye used to image the stamped polycation, PDAC, was 6-carboxyfluorescein (6-CF), which was purchased and used as received from Sigma. The dye was dissolved directly in 0.1 M NaOH; samples were imaged by dipping the substrates into the dye solution. The dye, which is negatively charged, selectively stained the positively charged PDAC surface. The dyed regions appear green when viewed with the fluorescence optical microscope, using a FITC filter.36 All fluorescence optical microscopy was done with a Zeiss Axiovert. Images were captured with a Hamamatsu C4742-95 digital camera and processed on a Macintosh G3 computer running Open Lab 2.0.2 software. All ellipsometry measurements were taken with a Gaertner Scientific Corporation ellipsometer, controlled by a Gateway 2000 computer running GEMP software with a He-Ne laser at 620 nm. After polyelectrolytes were stamped atop a multilayer platform, the topography of the stamped polyelectrolyte layer was observed using the tapping mode of a Digital Instruments Dimension 3000 atomic force microscope (AFM) with a silicon etched tip (TESP).

stamping onto polyelectrolyte multilayer substrates was demonstrated using 10(PDAC/SPS) bilayers adsorbed on glass slides and capped with a final layer of PAH. A single layer of polyelectrolyte was also used as a substrate; in this case, PAH was directly adsorbed on a gold-coated silicon wafer. These reflective samples were used for grazing angle FTIR studies. Propylaminosilane SAMs were used as substrates for the stability studies. In this case, propylaminosilane SAMs were formed on Si substrates by immersing piranha-cleaned Si substrates into a 2 mM ethanol solution of aminopropyltrimethoxysilane for 2 h. Microcontact Printing of Block Copolymer. The general procedure of polymer-on-polymer stamping (POPS) is shown in Figure 1. A 10 mM PS-b-PAA/THF solution (concentration based on the formula weight of the nominal repeat unit of styrene and acrylic acid) was used to ink untreated PDMS stamps molded from lithographically prepared masters.2 After evaporation of solvent, the PDMS stamp was briefly dried under a N2 stream and was brought into contact with the substrate for 10-15 min at room temperature. All stamped surfaces were then rinsed with ethanol to remove unbound or loosely bound excess polymer. The substrates used, as described above, include strong polyelectrolyte multilayers capped with PAH at pH 8.5, a single adsorbed monolayer of PAH on silicon, and an amino-functionalized SAM on a Si substrate. A PDMS stamp containing an array of 10 µm holes was used, and a water condensation image was immediately taken after the stamping process under an optical microscope.2 Characterization. Grazing angle FTIR spectra were obtained in single reflection mode using Digilab Fourier transform infrared spectrometer (Biorad, Cambridge, MA). The p-polarized light was incident at 80° relative to the surface normal of the substrate, and a mercury-cadmium-telluride (MCT) detector was used to detect the reflected light. A spectrum of a SAM of n-hexdecanethiolate-d33 on gold was taken as a reference.35 The buffer solutions used in stability tests were made according to the CRC Handbook of Chemistry and Physics (78th ed., 1997-1998) but diluted with deionized water to a final ionic strength equal to 10 mM. Accurate pH values were then measured with a pH meter after dilution. Potassium hydrogen phthalate (Aldrich) was used for the preparation of buffer solutions in the range of pH 2-5. Potassium dihydrogen phosphate (Aldrich) was used for the preparation of buffer solutions in the range of pH 7-10. Contact angles were measured on a Rame´-Hart goniometer (Rame´-Hart Inc., Mountain Lakes, NJ) equipped with a video-imaging system. Water drops were placed on at least three or more different locations on the surface in an ambient environment and advancing and receding contact angles were measured on both sides of each drop. The contacting water drops were advanced and retreated with an Electrapipet (Matrix Technologies, Lowell, MA) at approximately 2 µL/s to obtain advancing and receding angles, respectively. All reported values are the averaged values, and standard deviations typically ranged from 1 to 3°. Unless otherwise specified, values reported here are advancing contact angles. 3. Stamping of Polyelectrolytes on Charged Multilayer Surfaces. Substrate Preparation. The strong polyelectrolytes SPS and PDAC were used to form multilayer platforms on which solutions of the same polymers could be stamped. The platforms were built on COOH SAM covered gold substrates, or glass slides cleaned with a dilute Lysol/water mixture in a sonicator. To start the first bilayer, the slides were then immersed for 20 min in the PDAC solution (0.02 M PDAC, of MW 100000-200000, in Milli-Q water, with 0.1 M NaCl, filtered with a 0.22 µm filter). Following a 2-min rinse, the slides were placed into the SPS solution (0.01 M SPS, of MW 70 000, in Milli-Q water, with 0.1 M NaCl, filtered with a 0.22 µm filter) and allowed to sit for 20 min. They were rinsed a second time and sonicated for 3 min prior to repeating the procedure to make the next bilayer.25 Microcontact Printing of Polyelectrolytes. The PDMS stamp surfaces had to be made polar to increase their wettability to the polyelectrolyte solutions so that the stamps could be smoothly inked. Thus, clean stamps were placed in air plasma for 20 s

The polymer-on-polymer stamping technique is demonstrated here with two examples: stamping of block copolymers and stamping of charged polyelectrolyte homopolymers. In the first example described here, a block copolymer containing numerous weak acid groups (PAA) and a nonreactive, uncharged polymer block (polystyrene) is directly transferred to amino functional surfaces. The resulting surfaces were then examined as a function of stamping temperature and substrate pH pretreatment to determine optimal conditions for stamping. In this case, the poly(acrylic acid) block undergoes secondary and covalent interactions with the underlying amine surface. The nature of attachment of the polymer to the surface can also be based solely on ionic interactions; to demonstrate this effect, in the second portion of this paper we explore the direct transfer of a positively charged homopolymer to a negatively charged substrate, illustrating the universal nature of the polymer stamping approach, and the range of interactions which can be used to produce a chemically patterned polymer surface. Block Copolymer Stamping. Initial investigations of polymer-on-polymer stamping involved the transfer of alternating copolymers of maleic anhydride and poly(ethylene glycol)-functionalized methacrylates to the amino functional surface of a polyelectrolyte multilayer. In this case the basis of the functionalization was the use of an anchored polymer backbone with grafts of PEO oligomers extending from the backbone.31 The use of a block copolymer containing an anchoring block and a

(35) Laibinis, P. E.; Bain, C. D.; Nuzzo, R. G.; Whitesides, G. M. J. Phys. Chem. 1995, 99, 7663-7676.

(36) Caruso, F.; Lichtenfeld, H.; Donath, E.; Mohwald, H. Macromolecules 1999, 32, 2317-2328.

Results and Discussion

2610

Langmuir, Vol. 18, No. 7, 2002

Figure 2. FTIR of PS-PAA block copolymer stamped on an amino-functionalized surface transferred at different temperatures.

second, surface functional block, should also lead to a means of producing dense high molecular weight polymer layers on surfaces. A 10 mM solution of a polystyrenepoly(acrylic acid) diblock copolymer (PS-b-PAA) in tetrahydrofuran was used to ink a poly(dimethylsiloxane) (PDMS) stamp. After the excess solvent was removed from the stamp surface through nitrogen or air-drying, it was placed on the substrate and allowed to remain in contact for 10 min. The acid groups on the PAA block are bound via dipole-dipole, hydrogen bonding, and/or ionic interactions with the amino groups on the underlying substrate, as well as covalent bonding through the formation of amide groups. Samples were rinsed with ethanol to remove excess polymer, leaving a bound polymer monolayer approximately 6.0-8.0 nm in thickness. The effects of stamping temperature and substrate pH pretreatment on the nature of bonding of the PS-b-PAA block copolymer to the surface were examined with blank, featureless stamps which were made on a clean Si wafer. To investigate the effect of stamping temperature using FTIR, it was necessary to eliminate the IR absorption from the polyelectrolyte multilayer platform; this was done by using a single layer of PAH directly adsorbed onto a gold substrate from a 10 mM aqueous PAH solution (concentration based on repeat unit) at pH 8.6 overnight. The adsorption of a PAH layer was confirmed by ellipsometry (13 Å thick) and grazing angle FTIR. The substrate was then used as a stamping platform to avoid absorption from multiple polyelectrolyte layers in the FTIR spectra. PS-b-PAA was stamped onto the PAH layer at room temperature, at 100 °C and at 130 °C respectively by holding the substrate with the stamp in an oven at the appropriate temperature for 10 min. The stamp was then removed and grazing angle FTIR (GA-FTIR) spectroscopy was conducted. The results obtained at each stamping temperature are shown in Figure 2. The GA-FTIR spectra illustrate that the nature of bonding of PS-b-PAA on the polyamine surface changes with the temperature. As the stamping temperature is increased, the carbonyl stretch band shifts from 1745 cm-1, characteristic of the acid COOH, to 1671 cm-1, characteristic of the amide CONH. These measurements indicate that the acid groups in the PAA block of the PS-PAA block copolymer undergo condensation reactions with the primary amine groups in the underlying PAH film at higher temperatures. The differences in GA-FTIR suggest that block copolymer monolayer adsorption is based primarily on interactions

Jiang et al.

Figure 3. Effect of stamping temperature on the stability of stamped PS-PAA block copolymer: advancing contact angles of PS-PAA stamped on amino-functionalized surfaces at various stamping temperatures.

such as hydrogen bonding between amine and acid groups or acid/base interactions at low temperature, whereas at high temperature more stable covalent bonds form between the acid groups of the PAA blocks and the amino groups of the surface. Differences in the degree of adhesion and stability of the bound block copolymer layers as a function of the stamping temperature were measured with advancing contact angle measurements with water (see Figure 3, “as stamped”). Untreated aminosilane SAM surfaces with unprotonated primary amine groups were used as a stamping platform for contact angle stability studies in order to avoid unwanted effects due to desorption or other changes in the polyelectrolyte multilayer platform itself under different solvent conditions. For all samples examined, the surfaces were rinsed with ethanol following the stamping process and used immediately for contact angle measurements. Literature data for the advancing contact angle of polystyrene homopolymer with water range from 87° to 90°,37,38 and control measurements made on a 3 µm thick film of polystyrene annealed in air gave a contact angle of 92°. The contact printed surfaces examined here have slightly higher advancing contact angle values ranging from 95 to 100°; these higher advancing contact angles are due to surface roughness; these films also exhibited contact angle hysteresis (30-40°) that is consistent with rough surfaces, with receding angles below 80°. The thickness of the transferred layer for these samples was 6.1 ( 2.9 nm. Despite this issue, the qualitative nature of the films and their stability are well indicated by the relative differences observed in the advancing contact angle. The contact angle of poly(acrylic acid) is approximately 15-20°,32 much lower than that of polystyrene, and of the contact angles observed in Figure 3, indicating that in all cases the PS block is relatively well segregated from the underlying PAA block, presumably due to large adhesive secondary interactions as well as covalent bonding of the PAA to the amino groups on the platform. The water contact angle of as-stamped block copolymer indicate a continuous increase with increasing stamping temperature, confirming that the higher the stamping temperature, the greater the number of PAA segments bound on the amine surface due to the formation of (37) Wu, S. J. Polym. Sci., Part C 1971, 34, 19. (38) Kwok, D. Y.; Lam, C. N.; Li, A.; Zhu, K.; Wu, R.; Neumann, A. W. Polym. Eng. Sci. 1998, 38, 1675.

Polymer-on-Polymer Stamping

covalent amide bonds, as discussed above. Lower contact angles result from the presence of small numbers of unbound PAA segments which can access the surface. This result is consistent with the FTIR data collected as a function of temperature discussed earlier. The block copolymer samples stamped at high temperature (80 °C and higher) were significantly stable to solvent, and it was possible to rinse these samples with THF, which is actually a good solvent for the bulk PS-PAA block copolymer. When the high temperature stamped block copolymer surfaces are rinsed with THF, highly uniform polymer monolayers are obtained with an average thickness of 8.0 ( 0.9 nm. The films are more uniform than those rinsed with ethanol due to the more effective solvency of THF. The stability of the PS-b-PAA block copolymer films contact printed at different temperatures was tested by immersing each sample into each of four different solvent conditions: 0.01 M HCl (pH ) 2.0) for 10 min, 0.01 M NaOH (pH ) 12.0) for 10 min, deionized water for 20 h, and THF for 10 min. These solvent systems present competing hydrogen bond and polar interactions that can weaken secondary interactions between the PAA block and the amine surface. Following immersion in a given solution or solvent, the surface was dried to remove excess solvent, and the contact angle with water was measured immediately afterward and compared to the contact angle measured before the stability test. These results are shown in Figure 3; it is important to note that the x-axis represents the temperature at which the polymer was stamped. All solvent systems were kept at room temperature. In general, it was found that exposure to aqueous solutions or polar solvents such as THF leads to slightly lower contact angles under all stamping temperatures, indicating that secondary interactions play a role in adhesion of the layer under all conditions. This increase in surface energy is due to rearrangements of the block copolymer on the surface that exposes hydrophilic acid groups to the surface. It is also possible for solvents such as THF to remove some of the adsorbed layer through dissolution. The introduction of covalent bonds via amidation at high temperature prevents rearrangements and dissolution; the result is a more stable monolayer, as indicated by the increase in contact angle with increasing stamping temperature for each solvent condition shown (0.01 M HCl, 0.01 M NaOH, pure deionized water, pure THF). In earlier studies, we had found that covalent bonds can be encouraged in the stamping of an anhydride-vinyl ether graft copolymer by treating the surface prior to stamping with high pH aqueous solution.31 The amino groups are most reactive in their unprotonated form and can more readily undergo condensation with acid or anhydride groups to form amide bonds. The stability results discussed earlier in Figure 3 were done on stamped neutral surfaces containing a high density of reactive unprotonated primary amines. To examine the effect of pH pretreatment of the platform surface prior to stamping the PS-PAA block copolymer, aminosilane SAM substrates were immersed into 10 mM concentrations of pH 2.5, pH 5, pH 7, and pH 10 buffer solutions for 5 min and then dried under a dry N2 stream. The PS-b-PAA block copolymer was stamped on the aminosilane SAM at room temperature for 10 min, and contact angle measurements were performed with water on the stamped surface. Figure 4 contains data on PS-PAA block copolymers printed on platforms that were pretreated with different buffer solutions. The contact angle of the transferred block copolymer prior to any solvent exposure (Figure 4, “as

Langmuir, Vol. 18, No. 7, 2002 2611

Figure 4. Effect of pH pretreatment on the stability of stamped PS-PAA block copolymer: advancing contact angles of PSPAA stamped on amino-functionalized surfaces with various pH pretreatments.

stamped” values) was only 60° for low and intermediate pretreatment pH values, indicating the presence of many free acid segments from the PAA block that were accessible to the surface. It was not until a pretreatment pH of 10 was used, above the pKa of the primary amine groups on the surface, that the “as stamped” contact angle increased to greater than 85°, indicating the formation of a stable and segregated monolayer of polystyrene on the top surface. It is presumed that the more reactive amino groups obtained on the surface at pH 10 form larger numbers of covalent bonds between the PAA block and the amine surface. On observation of the contact angles of stamped surfaces following pretreatment, it is apparent that block copolymers stamped onto surfaces at pH 2.5-7 undergo increases in contact angle following some of the solvent treatments. Several possible adhesive interactions can dominate in the stamping process: ionic interactions between oppositely charged surface amine groups NH3+ and carboxylate groups COO- from PAA due to acid/base exchange, simple polar-polar interactions between NH2/NH3+ and COOH/COO-, hydrogen bonding, and ultimately covalent bonds. Samples pretreated to protonate the amine surfaces with acid solutions are stamped with a dry PS-PAA block copolymer layer. It appears that the acid/base or ionic interactions between the two surfaces are not optimized under these conditions; however, exposure to any aqueous solution, basic or acidic, leads to some increase in adhesion of the PAA block to the amine surface, as indicated by the higher contact angles following solvent stability treatments. The aqueous medium provides the opportunity for effective acid/base exchange between acid and amine groups and displacement of small counterions on the pretreated protonated surface, allowing for a stronger bond to that surface. It is clear that rearrangements of the PAA block do occur under these conditions. It suggests that ionic interactions are important for stamping done on charged amine surfaces pretreated at low pH conditions, and may well dominate the adhesion in this case. The formation of some covalent bonds likely contributes to film stability when the substrate is pretreated using high pH aqueous solutions to create a neutral amine surface, as was utilized for the data in Figure 3. Although the extent of this reaction may be relatively small, even small degrees of covalent bonding at the surface can lead to much more stable monolayers. The formation of covalent bonds is also consistent with the findings of the stability test: the contact angles actually increased dramatically

2612

Langmuir, Vol. 18, No. 7, 2002

Jiang et al.

Figure 5. The hydrophobic PS surface contrasted with the relatively hydrophilic aminosilane SAM surface (a) and the PAH surface (b) in water condensation image. Water drops are condensed on polyamine surface.

after the samples were soaked in a basic 0.01 M NaOH, which has a pH of 12. To illustrate the ability to directly micropattern with the block copolymer, water condensation images were formed from the alternating hydrophobic/hydrophilic surface regions of a stamped PS-b-PAA monolayer on the amine or polyamine surface. This procedure was demonstrated earlier by Kumar et al.2 for patterned alkanethiolate SAMs on gold surfaces. Figure 5a illustrates a condensation figure formed from stamping of the PSPAA block copolymer at room temperature, onto an aminosilane SAM without any pretreatment. The pattern of hydrophobic versus hydrophilic regions delineated by the water droplets illustrates a very clean, well-defined pattern. The diameter of the circular features shown is 10 µm. To demonstrate the successful stamping of PS-PAA onto a polymer multilayer surface, a strong polyelectrolyte platform of SPS/PDAC was formed on glass slides, followed with a top polyamine layer of PAH adsorbed at pH ) 8.5 (shown in Figure 5b). In both cases, a clear image is observed under the microscope of water droplets condensed and pinned on the surface. The outer regions contain the hydrophobic PS surface functional groups, whereas the 10 µm dots are the unstamped, hydrophilic amino regions which are readily wet by water. These images illustrate that the transfer of the polymer to the surface was effective and reproducible, even at room temperature and using relatively short stamp contact times, and can be done to micrometer resolution using a PDMS patterned stamp. Polyelectrolyte Stamping. The interactions between acid and amine groups during the stamping process can be optimized by varying pH or stamping temperature to encourage the formation of covalent bonds. Strong oppositely charged polyelectrolytes such as SPS and PDAC are highly ionized over all or most of the pH range and will only undergo ionic interactions with each other. These systems are ideal for examining the transfer of charged polymers onto an oppositely charged substrate based solely on ionic interactions. In the study reported here, PDAC was stamped directly onto an outermost SPS layer of a PDAC/SPS multilayer film. The stamped regions were characterized using ellipsometry, AFM, and fluorescence microscopy, as will be discussed below. Characterization of Transferred Polyelectrolyte Films. Microcontact printing of polymer systems involves the physical transfer of the polymer to a functional surface; under most circumstances, the material transferred by

the stamp exceeds that of a single functional polymer monolayer. The excess polymer is then rinsed away with an appropriate solvent, leaving an adsorbed monolayer of the polymer on the surface. Similar procedure is used to transfer common low molar mass monolayer-forming systems such as silanes and alkanethiols using microcontact printing.2 It is the single functional polymer monolayer which is of interest for applications involving the chemical patterning of surfaces. The polymer film that is initially transferred during the stamping of polyelectrolytes on surfaces is particularly thick due to the large cohesive interactions between polymer chains and the viscous nature of the high molecular weight polymer. The excess polymer is easily rinsed away with water, leaving only the desired functional monolayer of interest strongly adsorbed to the surface. To determine the amount of material transferred during the stamping process, negatively charged silicon oxide substrates were directly stamped with the polyelectrolyte, PDAC. The thicknesses of the stamped area prior to rinsing ranged from 20 to 30 nm; the monolayer obtained after rinsing was 5.0 ( 0.6 nm based on measurements of multiple samples. It is clear from these numbers that the initial excess polymer layer has been thoroughly removed in the rinsing process, leaving behind a highly uniform monolayer. The statistical average thickness of the transferred monolayer is similar in scale to the average thickness of 2.0 nm observed for a single adsorbed polyelectrolyte monolayer in the layerby-layer adsorption process. This observation suggests that the transferred layer may be similar to that obtained upon adsorption from solution. This observation provides an interesting comparison between polyelectrolyte layers adsorbed from dilute solution and those adsorbed during the stamping process; this topic is a part of ongoing studies. These data are also consistent with AFM images taken of a stamped PDAC film. The AFM of the surface before and after rinsing is shown in Figure 6. Prior to rinsing, the thickness of the stamped polymer layer shown is about 16 nm on average; following the rinsing process, the remaining polymer film is only 3-4 nm in height, a number which is again consistent with the range of thickness observed in polyelectrolye multilayer adsorption. Interestingly, a clear and sharp image is obtained in both cases. The AFM is able to detect topographical differences in the film to image the patterned polymer monolayer; unfortunately, the surface roughness of the polyelectrolyte multilayer approaches and in many cases surpasses the

Polymer-on-Polymer Stamping

Langmuir, Vol. 18, No. 7, 2002 2613

Figure 6. AFM images and section analysis of a transferred PDAC layer on PDAC/SPS multiplayer surface before (a) and after (b) rinsing with DI water.

thickness of the transferred polymer monolayer, resulting in a great deal of noise in the final images and making the AFM a less effective tool in imaging the actual chemical pattern produced using this method. For this reason, fluorescent imaging has been used to observe the transferred polymer monolayers on the surface. To image regions of alternating positive and negative charge, stamped multilayer samples were stained with a negatively charged fluorescent dye, 6-CF (6-carboxyfluorescein, from Sigma),36 for a few seconds to no more than 5 min and sonicated for 2 min in water. They were then examined in a fluorescence optical microscope. The resulting prints produced were stable and could be viewed weeks later with no change in appearance. All of the fluorescence images discussed in the following sections are of printed regions which were thoroughly rinsed and sonicated after stamping, indicating the presence of alternating charge on the surface region due to the presence of a monolayer of adsorbed polymer. Optimization of Stamping Process. Details of the stamping optimization process given here include information on the role and range of solvent choice, concentration, and stamp contact times applicable to the microcontact printing of a simple polyelectrolyte system. To be able to transfer charged or highly polar inks to the surface, it was necessary to treat the PDMS stamp with air plasma. Plasma times of 15 s or longer are sufficient to make the stamp wettable to aqueous solution, as determined by contact angle measurements of a blank PDMS surface (see Figure 7), which indicates a strong drop in contact angle of the PDMS surface at 15-20 s. The strong polyelectrolyte PDAC was stamped from dilute aqueous solutions and from water/ethanol mixtures at higher concentrations. Patterning can be achieved from very dilute polyelectrolyte solutions using aqueous inking solutions containing 0.02 M PDAC with 0.1 M NaCl added and 30 min of stamping time, provided that the PDMS stamp was plasma treated for 15 s. Similar results were observed with the stamping of negatively charged SPS on a PDAC surface of an PDAC/SPS film using a 0.01 M SPS aqueous solution with 0.1 M NaCl as ink, 30 s of air plasma

Figure 7. Contact angle measurements of a blank PDMS surface with air plasma at different periods of time.

for the PDMS stamp, and a 30 min stamping time (details not shown in this paper). PDAC was also successfully stamped from concentrated solutions in water/ethanol mixtures at much shorter stamping times. Several conditions were examined for optimization of stamping from these solutions. Ethanol is a promising solvent because it is at once polar and volatile, which suggests that it can easily solubilize PDAC, yet it evaporates rapidly from the stamp, preventing “bleed” during the stamping process. Neither pure ethanol, nor a 25/75 by volume ethanol-water mixture produced good coverage of the polyelectrolyte. Insufficient coverage can produce black areas with no transferred film or areas where only the edges of the pattern were transferred (rimming). This effect is shown in Figure 8a. Solutions made in either 50/50 ethanol/water or 75/25 ethanol/water mixtures performed much better, with the 75/25 mixture giving uniform transferred polymer films, with welldefined, clear edges. For most of these solutions, variations in coverage on the stamp during the inking process often produced streaks similar to brush strokes, as pictured in Figure 8b.

2614

Langmuir, Vol. 18, No. 7, 2002

Jiang et al.

Figure 8. Examples of defects which occur under nonideal stamping conditions: (a) Rimming can occur in areas of the stamp which were insufficiently inked (bare regions are also possible). (b) Streaking due to uneven application of the PDAC inking solution on the PDMS surface. Streaks are parallel to brush strokes used to apply ink to the stamp. (c) Surface cracks created after excessively long stamp contact times.

Figure 9. Fluorescent images of PDAC stamped onto PDAC/SPS multiplayer surface: (a) a close view image and (b) a wide angle image of a different sample stamped at optimal conditions.

A wide variety of stamping contact times were trieds ranging from a few seconds to an hour. At longer stamping times, the stamp tended to adhere to the platform, making it difficult to remove. Due to this phenomenon, the resulting printed areas displayed cracked surface regions (see Figure 8c). Shorter stamping times reduced the sticking, as did lighter coatings of ink. At very short stamp times (a few seconds), no PDAC was transferred to the platform. These results indicate that there is an optimal contact time for printing; for PDAC/ethanol solutions, optimal times were found at 30 s to 1 min, which resulted in good prints without great difficulty in removing the stamp. A number of concentrations of PDAC were also attempted using the ethanol/water solvent mixtures. Low concentrations (0.025 M) produced poor or no transferred prints at all. Moderate concentrations (0.1 M) performed well, but high concentrations (0.25 M) performed best, giving highly uniform stamped regions over large areas. On the basis of the variables discussed above, the optimal stamping condition for PDAC was determined to be a 0.25 M solution of PDAC in a 75/25 ethanol-water mixture, stamped for 1 min. This set of conditions is markedly different from the successful aqueous stamping

conditions, which work best at dilute concentrations and longer stamping times. Images are shown for samples stamped using optimal conditions from ethanol/water mixtures in Figure 9. The presence of an alternating positive/negative pattern is made clear by the presence of the green negatively charged dye on the positive PDAC regions. The dark black regions are the underlying SPS layer, which is not stained by the dye because of electrostatic repulsion. It is clear that there is no bleed or unwanted transfer of PDAC in the SPS regions of the pattern, indicating a clean pattern transfer. A uniform layer such as the one shown can be created over large areas. We have successfully patterned micrometer-sized features over approximately a centimeter square area; the possibility of patterning over large areas is therefore reasonable using this approach. Conclusions A new approach to patterning surfaces, polymer-onpolymer stamping, was developed to expand the techniques of microcontact printing to a number of different surfaces. This method involves the microcontact printing of graft and block copolymers on a range of materials, including common plastic substrates and the surfaces of

Polymer-on-Polymer Stamping

polyelectrolyte multilayers, to obtain functional patterned surfaces of multiple level layer-by-layer thin films. In this approach, the top surfaces of polymer films are chemically patterned through the direct stamping of functional polymers on the surface. The resulting pattern may then be used as a template for the further deposition of materials on the surface. A PS-PAA block copolymer was stamped onto amino-functionalized surfaces based on ionic, polar, and covalent bonding between acid and amine groups. The stability of the resulting polymer monolayer varied with pH treatment and temperature of stamping. Alternating positively and negatively charged regions were also created by stamping positively charged polymers onto negatively charged surfaces. Because a number of existing polymers may be used for this technique, many new functional surfaces can be achieved through the use of block copolymers or graft copolymers. This approach is commercially viable, for the modification of flexible film

Langmuir, Vol. 18, No. 7, 2002 2615

using roll-by-roll processing does not require substantive surface pretreatments in the case of many polymer and plastic substrates. Acknowledgment. The authors gratefully thank Professor Michael F. Rubner for his inspiration, insight, and encouragement, as well as Jeeyoung Choi for helpful discussions and interactions. We also thank Yas Hashimura for his experimental work in this project, and the Griffith-Lauffenburger Lab at MIT for the use of fluorescence microscope. Funding for this project was provided by the Office of Naval Research Polymers Program on Grant No. N00014-96-1-0789. Funding and instrumentation facilities were also provided by the MIT Center for Materials Science and Engineering MRSEC program of the National Science Foundation, DMR-9400334. LA011098D