Enhanced Reversible Adhesion of Dopamine Methacrylamide-Coated

May 20, 2009 - In this work, we take previously developed gecko-foot-hair-inspired elastomer microfiber arrays with film-terminated and mushroom-shape...
0 downloads 0 Views 3MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Enhanced Reversible Adhesion of Dopamine Methacrylamide-Coated Elastomer Microfibrillar Structures under Wet Conditions Paul Glass,† Hoyong Chung,‡ Newell R. Washburn,*,†,‡ and Metin Sitti*,†,§ †

Department of Biomedical Engineering, ‡Department of Chemistry and §Department of Mechanical Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 15213 Received March 15, 2009. Revised Manuscript Received May 7, 2009

In this work, we take previously developed gecko-foot-hair-inspired elastomer microfiber arrays with film-terminated and mushroom-shaped tips that have demonstrated enhanced adhesion with respect to unpatterned materials under dry conditions and coat them with synthetic DOPA-containing mussel-inspired polymers to enhance adhesion repeatedly in fully submerged wet environments. A new protocol for the development of this hybrid patterned, coated adhesive, which is suitable for use in contact with both wet and dry nonflat surfaces, is described. The experimental evaluation of repeatable adhesion under both wet and dry conditions for these materials is described and compared with unpatterned and/or uncoated materials. Macroscale reversible fibrillar adhesion enhancement on a nonflat, smooth glass surface when compared with unpatterned materials under fully submerged conditions is demonstrated with no suction effect.

Introduction Developing new adhesive materials that robustly and repeatedly attach and detach from wet surfaces is a major challenge. To enhance the adhesion of synthetic materials to dry surfaces, researchers have studied and developed materials inspired by systems found in nature, such as the micro/nanofibrillar structures on the feet of geckos.1-10 However, these fibrillar materials tend not to work when wetted with water,11,12 and none of these materials have demonstrated considerable adhesion under water. Only recently, Varenberg and Gorb13 demonstrated enhanced underwater adhesion of mushroom-shaped microfibrillar structures on flat surfaces using a suction effect. Additional research has also been conducted to study the wet adhesive ability of mussels, whose adhesive strength is attributed to cross-linking between polymer chains of individual adhesive proteins found in mussel holdfasts.14-16 This cross-linking of polyphenolic mussel proteins primarily requires a catecholic precursor (3,4-dihydroxyphenyl-L-alanine (DOPA)) and the presence of a catechol *To whom correspondence should be addressed. E-mail: washburn@ andrew.cmu.edu, [email protected]. (1) Autumn, K.; Liang, Y.; Hsieh, T.; Zesch, W.; Chan, W. P.; Kenny, T.; Fearing, R.; Full, R. J. Nature 2000, 405, 681. (2) Autumn, K.; Sitti, M.; Liang, Y. A.; Peattie, A. M.; Hansen, W. R.; Sponberg, S.; Kenny, T.; Fearing, R.; Israelachvili, J. N.; Full, R. J. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 12252. (3) Arzt, E.; Gorb, S.; Spolenak, R. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10603. (4) Geim, A. K.; Dubonos, S. V.; Grigorieva, I. V.; Novoselov, K. S.; Zhukov, A. A.; Shapoval, S. Y. Nat. Mater. 2003, 2, 461. (5) Sitti, M.; Fearing, R. S. J. Adhes. Sci. Technol. 2003, 17, 1055. (6) Crosby, A. J.; Hageman, M.; Duncan, A. Langmuir 2005, 21, 11738. (7) Kim, S.; Sitti, M. Appl. Phys. Lett. 2006, 89, 261911. (8) Glassmaker, N. J.; Jagota, A.; Hui, C. Y.; Noderer, W. L.; Chaudhury, M. K. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 10786. (9) Aksak, B.; Murphy, M.; Sitti, M. Langmuir 2007, 23, 3322. (10) Murphy, M.; Aksak, B.; Sitti, M. J. Adhes. Sci. Technol. 2007, 21, 1281. (11) Huber, G.; Mantz, H.; Spolenak, R.; Mecke, K.; Jacobs, K.; Gorb, S. N.; Arzt, E. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 16293. (12) Sun, W.; Neuzil, P.; Kustandi, T. S.; Oh, S.; Samper, V. D. Biophys. J. 2005, 89, L14. (13) Varenberg, M.; Gorb, S. J. R. Soc. Interface 2008, 5, 383. (14) Waite, J. H. Comp. Biochem. Physiol., Part B 1990, 97, 19. (15) Yu, M.; Hwang, J.; Deming, T. J. J. Am. Chem. Soc. 1999, 121, 5825. (16) Yu, M.; Deming, T. J. Macromolecules 1998, 31, 4739.

Langmuir 2009, 25(12), 6607–6612

oxidase.14 The reactive oxidized form of DOPA, quinone, is thought to provide the mussel’s water-resistant adhesive ability.15,16 Because of the water-resistant adhesive properties of DOPA residues that allow for rapid and strong adhesion to both organic and inorganic surfaces, other research has been conducted to mimic the chemistry of DOPA with synthetic polymers.15,17-20 It has recently been shown that by coating geckoinspired nanopillar surfaces with a synthetic mussel-inspired adhesive containing DOPA it is possible to generate reversible adhesion to both organic and inorganic substrates, even under wet conditions.21 However, whereas these coated materials demonstrated adhesion enhancement on the nanoscale, no macroscale performance was evaluated and no systematic enhancement was reported when compared to unpatterned materials. In this work, we take previously developed gecko-foot-hair-inspired elastomer microfiber arrays with film-terminated8 and mushroom-shaped tips,7,10 which have demonstrated enhanced adhesion with respect to unpatterned materials under dry conditions, and coat them with these same synthetic DOPA-containing mussel-inspired polymers to enhance adhesion repeatedly in fully submerged wet environments. A new protocol for the development of this hybrid patterned and coated adhesive, which is suitable for use in contact with both wet and dry nonflat surfaces, is described. The experimental evaluation of repeatable adhesion under both wet and dry conditions for these materials is described and compared with unpatterned and/or uncoated materials. In this work, macroscale reversible adhesion enhancement, as opposed to previous microscale AFM-based adhesion studies,4,21 on a nonflat (hemispherical) smooth glass surface is observed when compared with unpatterned materials under fully submerged conditions and is accomplished with no suction effect. (17) Sever, M. J.; Weisser, J. T.; Monahan, J.; Srinivasan, S.; Wilker, J. J. Angew. Chem., Int. Ed. 2004, 43, 448. (18) Sun, C.; Waite, J. H. J. Biol. Chem. 2005, 280, 39322. (19) Huang, K.; Lee, B.; Ingram, D.; Messersmith, P. B. Biomacromolecules 2002, 3, 397. (20) Lee, B. P.; Chao, C.-Y.; Nunalee, F. N.; Motan, E.; Shull, K. R. Messersmith, P. B. Macromolecules 2006, 39, 1740. (21) Lee, H.; Lee, B. P.; Messersmith, P. B. Nature 2007, 448, 338.

Published on Web 05/20/2009

DOI: 10.1021/la9009114

6607

Letter

Results and Discussion Coated adhesives were prepared first by synthesizing dopamine methacrylamide (DMA) and then from this synthesizing poly(dopamine methacrylate-co-2-methoxyethyl acrylate) (p(DMAco-MEA)). The synthesis of DMA and p(DMA-co-MEA) is illustrated in Figure 1. Finally, patterned polyurethane samples were prepared, and a stamping process was conducted to transfer p(DMA-co-MEA) onto the fiber array. DMA was prepared according to a modified method based on a previously reported technique21 under moderately basic conditions from the reaction of 3,4-dihydroxyphenethylamine hydrochloride and methacrylate anhydride in an aqueous solution of sodium borate and sodium bicarbonate. Aqueous sodium borate solution, the reaction medium, protects the dopamine moiety by forming the borate ester.22 Synthesized DMA was not water-soluble, so N,N-dimethylformamide (DMF) was used as a reaction medium to perform copolymerization. The synthesized DMA monomer was a solid pale-gray powder prepared in a yield of 44%. DMA was copolymerized with 2-methoxyethyl acrylate (MEA) via thermally initiated free radical polymerization. Proton nuclear magnetic resonance (1H NMR) spectral analysis on samples of the synthesized DMA and p(DMA-co-MEA) was performed to confirm that the resulting materials had the desired chemical structure. In this case, the synthesized polymer had a 1:5.4 DMA/ MEA ratio. Catechol contents for our sample can be converted to a weight percentage, 24%, which is almost the same as previously published data.21 It has been shown that a higher catechol content in the synthetic polymer gives stronger reversible adhesion in water.20 A 50 mg/mL solution of p(DMA-co-MEA) was prepared by dissolving the polymer in methylene chloride. Methylene chloride was selected as a solvent for deposition on the pillars because its high volatility allowed for rapid evaporation and quick pillar coating. The process diagram for creating the DOPA-containing polymer-coated, patterned fiber arrays is illustrated in Figure 2. Arrays of polyurethane (ST-1060, BJB Enterprises) pillars with approximately 40 μm fiber diameter, 80 μm spacing between adjacent fiber edges, and 100 μm fiber height were manufactured using a previously published lithographic and molding technique.9 The p(DMA-co-MEA) solution was manually spread onto a flat surface and the solvent was allowed to evaporate. p(DMAco-MEA) was applied to the fiber arrays by spin-coating a small amount of liquid polyurethane onto a second surface at 4000 rpm for 30 s. The fiber arrays were dipped into the liquid polyurethane layer (as shown in Figure 2a), removed such that a small amount of liquid ST-1060 coated each tip (as shown in Figure 2b), and then placed tip down on the p(DMA-co-MEA)-coated surface (as shown in Figure 2c). After the tips had cured, the arrays were peeled off of the surface, pulling a layer of DOPA-containing polymer from the substrate onto the fiber tips, as seen in Figure 2d. Attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) was used to confirm the existence of p(DMA-co-MEA) on the coated fiber array, which is discussed in more detail in the Supporting Information section. Because the ST-1060 polyurethane does not have strong resistance to organic solvents, even brief exposure to organic solvents resulted in significant deformation of the patterned structures from their original geometry. Therefore, previously reported dip coating in an ethanol solution21 was not a suitable method for polyurethane coating, necessitating this more complex multistep process. (22) Yamamoto, Y.; Miyahera, Y. U.S. Patent 5,098,999, 1992.

6608 DOI: 10.1021/la9009114

Figure 1. Illustration of the synthesis of dopamine methacrylamide (DMA) and poly(dopamine methacrylate-co-2-methoxyethyl acrylate) (p(DMA-co-MEA)).

Figure 2. Process diagram for material fabrication. (a) Arrays of untreated pillars are brought into contact with a polystyrene surface coated with a thin layer of liquid polyurethane. (b) The array is retracted, leaving a small amount of polyurethane on each pillar tip. (c) The array is stamped onto a second polystyrene surface coated with p(DMA-co-MEA). (d) After the polyurethane has cured, the array is peeled away from the surface, pulling a layer of p(DMA-co-MEA) with it. Uncoated arrays of fibers with mushroom-shaped tips are obtained by allowing dipped fibers to cure on an uncoated, low-surface-energy surface after step b, before peeling.

Four control samples were also prepared to evaluate the effects of both coating and patterning on adhesion: (1) flat uncoated samples of ST-1060; (2) flat ST-1060 samples coated with p(DMA-co-MEA) using the above process; (3) identically patterned, uncoated ST-1060 pillar arrays with mushroom-shaped tips prepared according to a previously described process;10 and (4) identically patterned, uncoated ST-1060 pillar arrays with film-terminated tips prepared according to a separate, previously described process.8 Scanning electron microscopy (SEM) images of the patterned samples are shown in Figure 3. Uncoated ST-1060 pillar arrays with film-terminated tips are shown in Figure 3a, uncoated ST-1060 pillar arrays with mushroomshaped tips are shown in Figure 3b, and the resulting samples treated through the described stamping process are shown in Figure 3c. For the coated structures, p(DMA-co-MEA) lines the top surface of the 14-μm-thick membrane attached to the tips of our fibers whereas the film-terminated fiber array has a 26-μmthick ST-1060 membrane attached to the fiber tips. The adhesion of these samples under both wet and dry conditions was evaluated in contact with a 6-mm-diameter glass Langmuir 2009, 25(12), 6607–6612

Letter

Figure 3. Scanning electron microscopy images of fabricated fibrillar adhesive samples. (a) Untreated array of 40-μm-diameter polyurethane fibers with membrane-terminated tips. (b) Untreated array of 40-μm-diameter polyurethane fibers with mushroom-shaped tips. (c) Array of identically patterned polyurethane fibers treated by stamping the uncured fiber tips on a p(DMA-co-MEA) layer and peeling after curing.

Figure 4. Experimentally determined adhesion of dry fibrillar and p(DMA-co-MEA) stamped materials with a 6 mm glass hemisphere for increasing preloads. Both coated and uncoated patterned materials demonstrate approximately a 2-fold adhesion enhancement over flat materials.

Figure 5. Experimentally determined adhesion of fully submerged patterned and p(DMA-co-MEA) stamped materials with a 6 mm glass hemisphere for increasing preloads. Coated materials of filmterminated fiber arrays demonstrate as much as a 23-fold adhesion enhancement over untreated materials.

hemisphere using a custom force measurement setup. A hemisphere is used in these experiments because it represents a special type of nonflat surface with a well-defined roughness and is also immune to misalignment problems during testing. Each sample was tested for 10 different preloads between 5 and 50 mN. To evaluate the repeatability of adhesion under wet conditions and to understand the underlying adhesion mechanism of these materials more clearly, the coated pattered sample was subjected to 120 consecutive test cycles at the same location under 50 mN preload forces to observe whether deterioration in the adhesive performance occurred. After waiting for 20 min, we performed an additional 20 test cycles to try to identify any effects due to possible relaxation of the materials over time. Adhesion of the four samples using increasing preloads in wet and dry environments is shown in Figures 4 and 5, respectively. Under dry conditions, the coated, patterned sample and uncoated, mushroom-terminated array demonstrated similar

(approximately 2-fold) adhesion enhancements over the other samples because of the geometric-patterning-based enhancement that has been widely studied. No enhancement was observed for the uncoated, film-terminated sample, likely because of the thickness of the terminal membrane, which did not efficiently prevent crack propagation during adhesion pull off. For wet testing, the enhancement of the p(DMA-co-MEA)-coated materials was much more pronounced. All uncoated samples demonstrated minimal adhesion over the range of preload forces measured. Flat coated samples demonstrated as much as a 7-fold enhancement over uncoated samples, whereas patterned, coated samples demonstrated as much as a 23-fold adhesion enhancement over uncoated materials, overcoming the limitations of polyurethanebased patterned adhesives for submerged applications. The inverted microscope setup allows for direct imaging of fibrillar contact with the hemisphere during testing. Figure 6 illustrates typical force-distance curves at 50 mN preload for

Langmuir 2009, 25(12), 6607–6612

DOI: 10.1021/la9009114

6609

Letter

Figure 6. Force-distance data at 50 mN preload for flat untreated and p(DMA-co-MEA)-treated film-terminated samples under submerged conditions during adhesion testing with a glass hemisphere. Inset photographs have been postprocessed for clarity and correspond to the area of contact of the patterned sample with the hemisphere at the points indicated on the force-distance curve.

a flat untreated sample as well as for a p(DMA-co-MEA)-treated film-terminated sample under submerged conditions. Microscope images illustrate the area of contact during the approach (Figure 6A,B) and retract (Figure 6C-F) cycles of patterned, treated materials testing. By comparing the adhesion hysteresis, the integrated area between the approach and retract curves during one test cycle, it is clear that the film-terminated coated samples significantly enhance the effective interface toughness with respect to flat untreated materials. Complete data for the adhesion hysteresis for all samples under all conditions are provided in the Supporting Information section. Image postprocessing was performed on these photographs to clarify the interaction of the material with the hemisphere during testing. In this processing, pixels appear darker if the point they represent is deformed from its original noncontact state. During the approach cycle, the area of contact increases as the glass hemisphere is indented into the material until the desired compressive preload force is met (Figure 6B). During retraction, the hemisphere is moved away from the material surface, resulting in interfacial forces that eventually become tensile. The maximum contact area stays constant through Figure 6D, as reported previously for this type of structure,8 because of the resistance to crack propagation of the film-terminated fibers. This crack trapping can be seen as a darker ring in Figure 6E, which is a result of highly deformed edges of the terminal membrane maintaining 6610 DOI: 10.1021/la9009114

contact with the hemisphere during retraction. In Figure 6F, after pull-off, some areas of the membrane are slightly wrinkled from their initial state, resulting in some darkening of the previous contact area, even after testing was completed. We attribute such changes in the terminal film appearance to the loading and deformation to which the sample was subjected. The observed submerged wet adhesion of patterned coated samples is due to the intrinsic adhesive properties of p(DMA-coMEA), the geometry of the microstructured materials, and the properties of the materials used. The adhesion mechanism of DOPA-based polymers and proteins toward surfaces has been investigated extensively. Lee et al. proposed that the adhesion can be caused by a coordination bond with an oxide surface and covalent bond formation with an amine surface via a Michaeltype addition reaction.23 Yamamoto et al. suggested that the adhesion mechanism is due to mixed hydrophilic and hydrophobic interactions24 and postulated that the hydrophilic side chains include hydroxyl and amino groups and can form strong hydrogen bonds with hydrophilic surfaces such as glass. Likewise, the hydrophobic part of polymer chains such as the methyl, ethylene, and benzene groups can selectively interact with hydrophobic (23) Lee, H.; Scherer, N. F.; Messersmith, P. B. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 12999. (24) Yamamoto, H.; Sakai, Y.; Ohkawa, K. Biomacromolecules 2000, 1, 543.

Langmuir 2009, 25(12), 6607–6612

Letter

Figure 7. Experimentally determined adhesion repeatability of fully submerged patterned and p(DMA-co-MEA)-coated adhesives with film-terminated tips at 50 mN preload force. (A) A 12-μm-thick terminal film demonstrates tremendous initial adhesive ability but rapid deterioration. (B) A 48-μm-thick terminal film demonstrates moderate initial adhesive ability but gradual enhancement with repeated loading. Inset photographs are the side-view optical microscope images of the corresponding fibrillar sample cross-section.

surfaces and present the hydrophilic components to an aqueous environement. Moreover, the cohesive strength of DOPA adhesion can be enhanced through interchain cross-linking.25,26 These investigations suggest that DOPA-containing materials provide diverse mechanisms for adhering to a broad range of surfaces. Whereas the enhancement due to the p(DMA-co-MEA) coating is clear, an additional enhancement is observed as a result of micropatterning, where the samples that are both patterned and coated demonstrate the highest adhesion under both wet and dry conditions, which is shown in Figures 3 and 4, respectively. The importance of the array geometry, especially the thickness of the terminal p(DMA-co-MEA) membrane, must be considered and (25) Burzio, L. A.; Waite, J. H. Biochemistry 2000, 39, 11147. (26) Holten-Andersen, N.; Waite, J. H. J. Dent. Res. 2008, 87, 701.

Langmuir 2009, 25(12), 6607–6612

has been previously studied.8 This thickness helps determine the extent to which crack-trapping is the dominant factor in explaining adhesive performance and subsequent material deterioration over time or whether rate- or time-dependent material behavior is the driving factor in explaining adhesion performance. The importance of membrane thickness is illustrated by the contrasting repeatability results for samples with two different p(DMAco-MEA) membrane thicknesses, as shown in Figure 7. For relatively thin membranes, such as the 12-μm-thick membrane in Figure 7A, adhesion performance on the first few cycles is extremely high. In this case, the crack-trapping nature of the filmterminated geometry dominates the material behavior.8 However, because of the thinness of the terminal film, damage to the coated membrane can occur over repeated testing, resulting in rapid deterioration of the adhesion performance. This observed DOI: 10.1021/la9009114

6611

Letter

deterioration is consistent with several previous studies where highly adhesive but somewhat fragile structures demonstrate reduced performance over repeated testing.27,28 Whereas the thicker 48 μm film observed for the sample shown in Figure 7B is more robust and does not display any performance deterioration over time, the thickness of the film reduces the crack-trapping ability of the array, diminishing the adhesion force possible for the first loading cycles. However, with repeated testing, the bulk effects of the viscoelastic polyurethane backing layers and fibers demonstrate a mild increase in adhesive performance, overcoming any potential damage to this more robust terminal membrane. One possible explanation for this increase in force is a change in the polymer molecular structure during loading and unloading, resulting in fibers with higher compliance in later test cycles. This would result in greater fiber elongation before pull-off (Figure 6B-E), which could explain the increase in measured adhesion force.9 When the sample was left for 30 min before testing resumed, the adhesive performance of this sample decreased considerably, possibly owing to an additional change in the polymer structure during this waiting time. It is important to note that despite the deterioration of the thinner film sample, the steady-state adhesion force value after repeated testing is still comparable to that of the thicker film sample, owing to the continued p(DMA-co-MEA) presence in the contact area. Future areas of research will require refining the manufacturing process, particularly the stamping steps, in order to manufacture larger, more uniform sample areas with optimal membrane thicknesses, preferably capable of maintaining high adhesion forces over many cylces. The materials described in (27) Davies, J.; Haq, S.; Hawke, T.; Sargent, J. P. Int. J. Adhes. Adhes. 2009, 9, 380. (28) Murphy, M. Ph.D. Thesis, Carnegie Mellon University, 2008.

6612 DOI: 10.1021/la9009114

this work demonstrate reversible fibrillar adhesion under wet conditions on the macroscale with respect to both unpatterned and uncoated materials on nonflat surfaces using intermolecular forces instead of suction forces. Versatile reversible materials capable of adhering to nonflat surfaces under both wet and dry conditions would be crucial for an array of applications ranging from medical devices such as wireless therapeutic capsule endoscopes,29 where reliable and noninvasive anchoring to a wetted tissue is required for a doctor to perform a clinically useful task, to mobile climbing robots30 that may need to move under both wet and dry conditions. Acknowledgment. We thank C. Lee and E. Tworkoski for their efforts in collecting preliminary adhesion data for treated materials and L. T. Sun for help with polymer preparation. We also thank B. Aksak, E. Cheung, S. Kim, M. Murphy, and the other CMU NanoRobotics Laboratory members for their valuable feedback and suggestions. Supporting Information Available: Additional information on the synthesis and proton nuclear magnetic resonance spectral analysis of dopamine methacrylamide and poly(dopamine methacrylate-co-2-methoxyethyl acrylate) (p(DMA-co-MEA)), macroscale adhesion measurement protocol, dry and wet adhesion hysteresis data of fibrillar and flat adhesive samples, and surface characterization of coated and uncoated samples with attenuated total reflection Fourier transform infrared spectroscopy. This material is available free of charge via the Internet at http:// pubs.acs.org. (29) Glass, P.; Cheung, E.; Sitti, M. IEEE T. Bio-Med. Eng. 2008, 55, 2759. (30) Murphy, M.; Sitti, M. IEEE-ASME T. Mech. 2007, 12, 330.

Langmuir 2009, 25(12), 6607–6612