6182
Langmuir 2008, 24, 6182-6188
Biomimetic Core–Shell Fibril for Enhanced Adhesion Matthew B. Havener,†,| Vincent Sica,†,| Tian Tang,‡ and Anand Jagota*,†,§ Bioengineering Program and Department of Chemical Engineering, Lehigh UniVersity, Bethlehem, PennsylVania 18015, and Department of Mechanical Engineering, UniVersity of Alberta, Edmonton, Alberta T6G 2G8, Canada ReceiVed October 19, 2007. ReVised Manuscript ReceiVed January 31, 2008 Fibrillar adhesive structures in nature are usually terminated by compliant plate-like elements that are critically important. We have fabricated a simple, model, core–shell fibrillar structure by coating an aluminum wire with (poly)dimethylsiloxane (PDMS). By partially etching the core metal, we obtain a compliant annular terminus. Measurements of the force required for this structure to detach from and slide against a glass substrate show that sliding is accommodated by a stick-slip mechanism and that substantial enhancement of adhesion can be achieved. A simple theoretical model, which is in good agreement with experimental data, shows that during the sticking phase the contact reduces in size and the mechanics of this process is controlled by the balance of energy release from the stretched PDMS and adhesion between it and the substrate.
1. Introduction Contact surfaces in lizards and insects consist almost universally of hierarchical fibrillar surfaces that promote adhesion.1–7 Fibrils generally have thin platelike terminal elements called spatulae that are important for their contact properties. They allow for increased compliance and a larger contact area with rough surfaces.7,8 There has been considerable recent work on design and fabrication of synthetic mimic structures.9–16 Several designs have been developed that show significant adhesion enhancement.11–16 Natural fibrillar systems usually exhibit greater * To whom correspondence should be addressed. Email:
[email protected]. † Bioengineering Program, Lehigh University. ‡ University of Alberta. § Department of Chemical Engineering, Lehigh University. | These authors contributed equally to the work.
(1) Hiller, U. J. Bombay Nat. Hist. Soc. 1976, 73, 278–282. (2) Irschick, D. J.; Austin, C. C.; Petren, K.; Fisher, R.; Losos, J. B.; Ellers, O. Biol. J. Linn. Soc. 1996, 59, 21–35. (3) Autumn, K.; Liang, Y. A.; Hsieh, S. T.; Zesch, W.; Chan, W.-P.; Kenny, T. W.; Fearing, R.; Full, R. J. Nature (London) 2000, 405, 681–685. (4) Tian, Y.; Pesika, N.; Zeng, H.; Rosenberg, K.; Zhao, B.; McGuiggan, P.; Autumn, K.; Israelachvili, J. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 19320– 19325. (5) Jagota, A.; Hui, C.-Y.; Glassmaker, N. J.; Tang, T. MRS Bull. 2007, 32, 492–495, See also other articles in this issue of the MRS Bulletin. (6) Eisner, T.; Aneshansley, D. J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 6568–6573. (7) Rizzo, N. W.; Gardner, K. H.; Walls, D. J.; Keiper-Hrynko, N. M.; Ganzke, T. S.; Hallahan, D. L. J. R. Soc. Interface 2006, 3, 441–451. (8) Persson, B. N. J.; Gorb, S. J. Chem. Phys. 2003, 119(21), 11437–11444. (9) Glassmaker, N. J.; Jagota, A.; Hui, C.-Y.; Kim, J. J. R. Soc. Interface 2004, 1, 23–33. (10) Hui, C.-Y.; Glassmaker, N. J.; Tang, T.; Jagota, A. J. R. Soc. Interface 2004, 1, 35–48. (11) Kim, S.; Sitti, M. Appl. Phy. Lett. 2006, 89, 261911–261913. (12) Gorb, S.; Varenberg, M.; Peressadko, A.; Tuma, J. J. R. Soc. Interface 2007, 4, 271–275. (13) Glassmaker, N. J.; Jagota, A.; Hui, C.-Y.; Noderer, W. L.; Chaudhury, M. K. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 10786–10791. (14) Noderer, W. L.; Shen, L.; Vajpayee, S.; Glassmaker, N. J.; Jagota, A.; Hui, C.-Y. Proc. R. Soc. London, Ser. A. 2007, 463, 2631–2654. (15) Ge, L.; Sethi, S.; Ci, L.; Ajayan, P. M.; Dhinojwala, A. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 10792–10795. (16) Majidi, C.; Groff, R. E.; Maeno, Y.; Schubert, B.; Baek, S.; Bush, B.; Maboudian, R.; Gravish, N.; Wilkinson, M.; Autumn, K.; Fearing, R. S. Phys. ReV. Lett. 2006, 97, 076103. (17) Shen, L.; Glassmaker, N. J.; Jagota, A.; Hui, C-Y. Soft Matter 2008, 4, 618-625. (18) Havener, M.; Tang, T.; Jagota, A. Proceedings of the Annual Meeting of the Adhesion Society; Adhesion Society: Blacksburg, VA, 2007. (19) Gao, H.; Yao, H. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 7851–7856. (20) Rand, C. J.; Crosby, A. J. Appl. Phys. Lett. 2007, 91, 261909.
performance in shear.2–4 For example, shear force resisting sliding of a gecko seta (about 200 µN) is considerably greater than the adhesive force needed for normal separation (about 20–40 µN).3,4 Only a few synthetic materials have been studied in this mode of deformation (e.g. refs 16 and 17). Additionally, there is a basic question of how terminal plates are formed, maintained, and possibly regenerated in nature, given that the fibrillar surface comprises of extracellular proteinaceous material. Micrographs of the fibril tips (Figure 1 in ref 5 and Figure 2 in ref 7) suggest that these are core–shell structures. If, in such a structure, the core were to be removed differentially, say by dissolution or abrasion, it would leave a thin plate-like shell that could remain open or might collapse. In either case it would provide a compliant terminal element. Because this process could be self-limiting due to putative collapse or kinetically limited removal of the core material, terminal elements would then be self-forming. Moreover, such a mechanism could be mimicked synthetically. In this study, using a simple model system, we demonstrate that a compliant terminal shell fabricated using a core–shell structure can be used for adhesion enhancement. Our system comprises aluminum wire coated with (poly)dimethylsiloxane (PDMS) cured to make a thin outer shell. An end is then cut off to expose the aluminum, which is etched away for a certain length with hydrochloric acid. The result is an annular section of PDMS at the terminus of the wire. We show that if such a structure is preloaded onto a flat substrate, it develops considerable resistance against shear. We have developed a simple model for the mechanics of adhesion of these structures under shear that is in good agreement with experiments.
2. Experimental Methods Model core–shell structures were made by coating PDMS (Sylgard 184, Dow Chemicals) onto aluminum wires. PDMS combined in a 10:1 ratio by weight of rubber base and cure was poured into centrifuge tubes and placed under vacuum for 30 min to eliminate gas bubbles. Four-centimeter lengths of aluminum wire 1 mm in diameter were cut and held in a fixture so that a 2 cm section of wire was suspended vertically. The fixture then dipped the 2 cm section of these wires into the silicone mixture for 1 min followed by a 5 min dripping period. The samples were then placed in an oven at 80 °C for 1 h to cure. After removal from the oven, the wires were cut to expose the bare aluminum at the end of the wire. Each sample was covered with
10.1021/la7032687 CCC: $40.75 2008 American Chemical Society Published on Web 05/22/2008
Biomimetic Core–Shell Fibril for Enhanced Adhesion
Langmuir, Vol. 24, No. 12, 2008 6183
Figure 1. (a) Optical micrograph of a terminal shell fabricated by coating an aluminum wire with PDMS and partially etching the wire. (b) Scanning electron micrograph of a coated, partially etched, and trimmed sample with shell thickness of about 25 µm.
Figure 2. Schematic drawing of the experimental apparatus for adhesion measurements.
Scotch tape, dipped in liquid nitrogen for 30 s, and cut using shears. A number of wires were held in another fixture and submerged in a bath of 30% HCl; the wires were held vertically with the cut end at the top. After removal from the acid the samples were rinsed with acetone and deionized water and carefully blown dry with N2; 20 min of exposure to HCl etched roughly 1.5 mm of aluminum, exposing an annular section of PDMS. This annular section was trimmed with a razor mounted on a vertical translation stage to make a clean edge. The control (lo ) 0 mm) was a section of wire coated with PDMS and cut the same way as the rest of the samples but with no etching of the aluminum wire. We will present data from two series of samples, (i) and (ii), of varying lengths (lo) in the range (0–2 mm). Figure 1a contains an optical micrograph of typical etched sample showing the annular PDMS shell and the conical etched core. Figure 1b is a scanning electron micrograph of a sample from set (i) in cross-section. For this set the shell thickness was approximately 25 µm. For set (ii), the shell thickness was about 50 µm but the wire had a somewhat noncircular cross section as a result of which shell thickness was not uniform. SEM micrographs showed the accumulation of debris from etching on the outside surface of the shell. For this reason, we needed a surface cleaning step prior to adhesion measurements. The samples were cleaned by pressing against Scotch tape and peeling them off to remove the particulate debris. Adhesion experiments were carried out using a custom-built apparatus shown schematically in Figure 2. The sample was clamped and held at an angle θ ) 25ο with respect to the horizontal orientation. This angle is similar in magnitude to that between the setal shaft and spatular film in geckos, about 30°.4 The clamp was attached to a DC motor; a load cell between the two measured the transmitted (shear) force. A glass slide was treated to produce a coating of a hydrophobic self-assembled monolayer as described previously.13 The slide was
Figure 3. (i) Measured horizontal (shear) and vertical (normal) forces during a typical experiment on a sample with lo ) 1.3 mm and d ) 254 µm. After an initial transition, the force traces are periodic, corresponding to stick-slip behavior. (ii) Optical micrographs of the contact region at selected points. In b we have added lines to indicate sample edges, which image poorly compared to the contact region.
removed, rinsed with isopropanol, and blown dry with nitrogen several times. This slide was held in a glass bracket that was connected via a second load cell to a motor that controlled vertical motion. The entire apparatus was mounted on an inverted optical microscope, and the experiment was viewed through an objective and recorded by a video camera. The bracket with the glass slide was lifted vertically until it made contact with the specimen. Formation of a contact was apparent as it resulted in the formation of a contact zone with distinctly different contrast as will be shown later. Beyond initial contact, the bracket was further lifted either until a predetermined vertical load was achieved or until a predetermined additional vertical displacement of d was imposed. After a rest period, typically with a duration of
6184 Langmuir, Vol. 24, No. 12, 2008
HaVener et al.
Figure 4. (i) Shear force measured during experiments on a sample with lo ) 1.3 mm and varying initial compression, d. (ii) Optical micrographs of the contact region following the initial compression but before application of shear displacement.
about 10 s, the sample was pulled in shear relative to the glass slide for a total displacement of 1 mm at a rate of 30 µm/s. Shear and normal force and displacement were recorded during the experiment using a LabView program along with a video of the contact region as viewed through the optical microscope. Subsequent analysis of the data and images was performed using MatLab. We performed three to five experiments for a given set of conditions, e.g., for a given length lo and precompression depth d.
3. Results and Discussion 3.1. Experimental Results. Figure 3i shows measured vertical and horizontal (shear) forces during a typical experiment on a sample from set (i) with lo )1.3 mm and d ) 254 µm. Figure 3ii shows optical micrographs of the contact between the sample and the glass slide at selected points indicated by letters in Figure 3i. As the bracket is raised, it brings the sample in contact with the glass slide. We measure a compressive vertical force (blue trace, positive numbers represent compression) and a shear force attempting to push the sample to the right (red trace, positive numbers represent a tensile force on the load cell). The vertical displacement is held fixed for about 10 s, and the first plateau in the data represents this stage of the experiment. After this “rest” period, with vertical displacement held fixed, a shear displacement is applied (point a). Figure 3ii,a shows that there is an approximately triangular contact between the sample and
Figure 5. Maximum shear force as a function of initial depth, d, for different values of lo.
the glass slide at this point. With increasing shear the vertical force reduces in magnitude, then changes sign and becomes tensile. Simultaneously, the horizontal shear force reduces in magnitude, also changes sign, and becomes tensile (as applied on the load cell). As this occurs, the contact reduces in size (Figure 3ii,a-c). Both force traces show a reduction in stiffness. At point c, there is a sudden load drop, which is accompanied
Biomimetic Core–Shell Fibril for Enhanced Adhesion
Figure 6. Geometry of the sample before compression (with original length of lo) and after the right end has been displaced vertically to depth d (with current length of l for the noncontact region). The horizontal positions of the ends of the sample are held fixed. During loading, a contact region is formed between the sample and the substrate, and its length is denoted by s. The inclined angle θ is assumed to be fixed during compression.
by unstable detachment of the sample, followed by slip, and reattachment (Figure 3ii,d). This stick-slip cycle repeats (Figure 3(iid-f)) until we halt the horizontal motion of the motor. On reattachment, the area is smaller than the initial contact. However, the maximum shear force remains about the same. Figure 4i shows shear force measurements during a series of experiments on the same sample (lo ) 1.3 mm) but with varying values of initial compression, d. Again, the first (negative) plateau in the data represents the compressive force on the horizontal load cell that develops during initial compression. Its magnitude generally increases with increasing d; for larger values in these and other experiments it saturates. Figure 4ii shows optical micrographs of the contact region for the eight cases shown in Figure 4i following application of initial compression but before any shear displacement is applied. The initial contact region increases systematically with increasing d. Figure 4i shows that both the maximum load sustained by the sample in shear and the period of the stick-slip cycles increase systematically with increasing d. Similar stick-slip adhesion in shear was observed on all samples. There is some variability in reattachment. For example, the trace corresponding to d ) 152.4 µm in Figure 4i shows less regular stick-slip. However, the contact geometry just before the first slip instability was always as shown in Figure 3ii,c,e.
Langmuir, Vol. 24, No. 12, 2008 6185
That is, the contact reduces in size as shear displacement is applied and, at peak load, the remaining contact is small and at the extreme end of the sample. We will take the maximum shear force as a measure of adhesion and will later present a simple model to explain our experimental results. Figure 5 shows that the maximum shear force increases with depth, d, for five different lengths, lo. We find that adhesion can be very strongly enhanced by application of an initial normal displacement. We also conducted a series of experiments on samples (ii). In this case, for each sample, we increased the vertical displacement until the vertical compressive force reached a value of 2.5 mN. We observed a systematic, approximately linear increase in maximum shear force with lo. For lo ) 1.25 mm, it reached a value exceeding 16 mN; for the control, lo ) 0, it was
