«UNIVERSITY OF CALIFORNIA Santa Barbara A Micro/Nano-Fabricated Gecko-Inspired Reversible Adhesive A Dissertation submitted in partial satisfaction of ...»
SHARPS structures consisting of a silicon dioxide platform supported by a single support have been fabricated. The structures were created through a unique process which allows three dimensional control with only a single lithographic step.
Platform geometry can be designed to control the geometry of the support pillar about the radial and long axis. Subsequent processing was used to further enhance the structure of the platform. Here sputtered titanium was used to alter the curvature of the platform and enhance adhesive properties. The adhesive properties were measured using a modified nanoindentation set-up. It was shown that the SHARPS structures can offer improved adhesion for non-planar surfaces.
The authors would like to thank Jesse Williams, Marco Aimi, Masa Rao, Adam Pyzyna, Jim Cooper and Eric Johnson. Without all their help this work would have not come to fruition. Funding for this work was provided by the University of California, Santa Barbara.
F. Figures Figure IV-1 Electron micrographs of various SHARPS structures (From Top and Left to Right) Square topped array, round top, slotted square, slotted round, branched finger, branched fine finger, radial meander, serpentine.
Figure IV-2 SHARPS fabrication process flow.
Figure IV-3 Optical micrographs showing pillar geometry along the long axis. (From left to right) Hexagonal, triangular, square, octagonal.
Figure IV-4 SHARPS structure with modified pillar geometry along the long axis.
Figure IV-5 (Top) surface profile taken in a Veeco Wyko N1100 of the structure in the bottom left image. (Bottom left) SHARPS structure after sputtering a half micrometer of titanium. (Bottom right) same structure after sputtering ~5 μm of titanium.
Figure IV-6 Nanoindenter bending stiffness experimental data and theoretical plot originating at the center of the SHARPS platform. (Top) Diamond cross section (Bottom) Square cross section.
Figure IV-7 (Top) before and after images of a probe pushing laterally on the structure. (Bottom left) Superposition of images illustrating deformation distance: distance between blue and yellow lines. (Bottom right) Schematic representation of 3-dimensional deformation.
Figure IV-8 SHARPS structures after sputtering ~5 μm titanium.
Figure IV-9 Comparison of adhesion vs. applied normal force between a polyamide bead and the substrate or a radial meander SHARPS structure.
V. A batch fabricated biomimetic dry adhesive
The fine hair adhesive system found in nature is capable of reversibly adhering to just about any surface. This dry adhesive, best demonstrated in the pad of the gecko, makes use of a multilevel conformal structure to greatly increase inelastic surface contact, enhancing short range interactions and producing significant amounts of attractive forces. Recent work has attempted to reproduce and test the terminal sub-micrometer “hairs” of the system. Here we report the first batch fabricated multi-scale conformal system to mimic nature’s dry adhesive. The approach makes use of massively parallel MEMS processing technology to produce 20-150 µm platforms, supported by single slender pillars, and coated with ~2 µm long, ~200 nm diameter, organic looking polymer nanorods, or “organorods.” To characterize the structures a new meso-scale nanoindenter adhesion test technique has been developed.
Experiments indicate significantly improved adhesion with the multi-scale system. Additional processing caused a hydrophilic to hydrophobic transformation of the surface and testing indicated further improvement in adhesion.
Used by insects and lizards (including flies, crickets, beetles, spiders, geckos and anoles) to climb wet or dry, vertical and even inverted surfaces, the fine hair adhesive system is an excellent example of convergent evolution in biology(1-4, 6-8, 54). Recently, much work has been done to better understand the science of the sticking of this fine-hair adhesive motif (6-8, 15, 55-57). Experimental evidence has shown that the adhesion is primarily due to short-range weak van der Waals interactions between the fine hairs on the adhering surface and the target surface(7).
In order for these “weak” forces to become significant the adhesive surface must create a large amount of intimate surface contact to the binding surface. In the case of the Gekko gecko, that can weigh up to 300 grams, this is achieved by a multilevel conformal system consisting of toes containing blood sinuses supporting rows of imbricated lamellae with densely packed keratinous setae approximately 100 µm in length, which split into finer 200 nm in diameter bristles (2). Each element of the system, from the toe to the terminal bristles, provides another scale of surface conformation, from the centimeter meso-scale (toe) down to the nano-scale (bristles). This inelastic surface conformation allows the gecko’s foot to create a large amount of surface contact, without introducing repulsive restoring force from the surface, producing a significant amount of surface adhesion through short-range interactions (15).
Prior work has focused on mimicking the terminal bristle component of the adhesive by fabricating arrays of polymeric nanorods on solid substrate(7, 11, 58).
While individual nanorods demonstrated expected amounts of adhesion, larger arrays failed to produce larger amounts of adhesion – unless removed from the substrate and placed on a compliant backing(11), showing the need for multi-scale compliance. Additionally, the arrays of nanorods showed reduced adhesion with use, due to bunching and contamination(11), suggesting the reason for the superhydrophobic nature of the gecko pad.
In the last two decades the emerging field of microelectromechanical systems (MEMS) has created a variety of micro-devices. These devices offer distinct advantages due to their small size, particularly in sensing. Microsensors have been developed for location determination, chemical sensing, mass sensing, pressure sensing, and inertial sensing(42, 44-48, 51, 52). One key issue is how to deploy these micro-devices. The ability to incorporate an adhesive into the fabrication process and have that adhesive stick to virtually any surface, in any environment, offers a significant technological advancement. Another key aspect in both microand nano-devices is the integration into complete device architectures. As device architectures are miniaturized and span multiple disciplines (biology, integrated circuits, MEMS, etc...) new strategies for assembly and integration will be needed.
The ability to pattern an adhesive with potentially sub-micrometer precision for chip recognition or self-assembly strategies will be a valuable tool.
In this work, a chip-scale batch fabricated multi-scale conformal system has been produced. The microelectromechanical systems (MEMS)-based approach allows for batch fabrication and chip integration of the adhesive. The multi-scale structures consist of arrays of organic-looking photoresist nanorods, “organorods”, approximately 2 μm tall and 50-200 nm in diameter, atop photolithographically defined 2 µm thick silicon dioxide platforms 100-150 μm on a side (Fig. 1). The platforms of varying geometries are supported by single high-aspect-ratio pillars down to 1 µm in diameter and heights up to ~50 µm (Fig. 2). The structures are fabricated out of 4-inch single crystal silicon wafers in the (100) orientation using standard bulk micromachining techniques(59). Using a wet oxidation process at 1100°C 2 μm of silicon dioxide is grown on the silicon wafer. To define the top platform geometry standard stepper photolithography in an i-line stepper is performed using a positive resist, Shipley SPR 220-7. The resist is then used as an etch mask in an ICP etcher with CHF3 chemistry to vertically etch through the silicon dioxide to the silicon. The exposed silicon is then etched using the Bosch™ process, also known as deep reactive ion etching (DRIE), where the plasma is cycled between a highly reactive SF6 gas and a hydrocarbon forming CF4 species, creating high-aspect-ratio vertical cavities. The depth of these cavities can be controlled depending on the desired final aspect ratio of the pillars. Directly following the deep etch an extended SF6 etch is used to isotropically etch the silicon. Since the reactive etch is significantly more selective to the silicon than the silicon dioxide (or photoresist) the platforms are undercut from all direction leaving behind only a single pillar in the middle (Fig. 2). By controlling the duration of the release it is possible to control the final size of the pillars supporting the platforms(59).
Following the platform and pillar fabrication, the photoresist surface (Shipley SPR 220-7, primarily composed of a diazoquinone ester and a phenolic novolak resin) of the platforms is transformed into organorods by placing it into an oxygen plasma with a 100 W bias for 5 minutes (Fig. 3), for a 3 cm x 3 cm piece.
Organorods have also been uniformly fabricated across a 100 mm silicon wafer with good uniformity. The bias creates an electric field gradient, which acting on the dielectric polymer induces a force large enough to overcome surface tension, causing the growth of vertical polymeric columns. Estimating the electric field across the photoresist to be 100 V/µm, a column size of around 1 µm is predicted(60), much larger than the 50-150 nm organorods. The reduced geometry of the organorods may be attributed to a bilayer of photoresist and hexamethyldisilazane (HMDS) decreasing the interfacial energy and shrinking the instability wavelength(61).
The use of a plasma induced electric field offers several advantages over a parallel plate induced field. Using a plasma to create the electric field requires no second electrode, and subsequently no precise gap control between the two plates.
From a fabrication standpoint plasma induced nanorod growth utilizes already established processing infrastructure. Adopted from the IC community, and available on most research campuses, plasma etching systems are convenient and established processing tools.
The most striking difference between the plasma grown nanorods in this work and the parallel plate grown nanorods done by Schäffer et al. is the reduced time it takes to form the nanostructures – 15 minutes versus 18 hours, respectively.
In essence the plasma method offers a greatly accelerated growth mechanism;
similar initial morphologies are seen before the formation of the organorods. A possible explanation for the greatly reduced formation time can be understood by examining the growth mechanism. Growth is caused by the electric field gradient exerting a force on the dielectric polymer. When this force exceeds the restoring force created by the surface tension of the polymer surface an undulated surface begins to appear and then transform into the polymeric pillars seen in figure 3 and ref (4). In the plasma growth method the oxygen plasma serves to reduce the surface tension of the photoresist by breaking C-H bonds at the surface and leaving dangling –OH bonds. These dangling bonds reduce the surface tension, leaving the dielectric force to dominate over the restoring force, and organorods are formed.
To change the organorod surface from hydrophilic to hydrophobic the samples are placed in a CF4 plasma for 30 seconds. This creates a fluorocarbon coating, increasing their size to ~350 nm, and altering their surface chemistry. The fluorocarbon coating leaves a -CF3 terminated surface, greatly reducing the surface energy. Combining this coating with the morphology of the organorod surface creates a lotus leaf effect(62), making the surface highly-hydrophobic with a water contact angle of 145° ± 2 (Fig. 4).
Adhesion testing was performed using a Hysitron Triboindenter® with a 3.175 mm spherical aluminum tip (RMS roughness of 0.5 µm over a 200 µm square). This non-ideal aluminum tip was chosen to better simulate an actual working environment for the adhesive, e.g. sensor deployment on an aircraft. To simulate a meso-scale adhesion incident, the “rough” tip was pressed into the test surfaces and withdrawn orthogonally from the surface at a constant rate. Operating in displacement control load versus displacement data was collected and analyzed from the nanoindenter to determine adhesion. Adhering surfaces would produce a distinctive pull-off behavior, where the unloading curve would make a sudden jump (Fig. 5). The adhesive force was taken to be the difference between the minimum value right before pull-off and the next stabilized point (the instrumentation requires a finite time to stabilize after the sudden pull-off event).
To the authors’ knowledge, this is the first time a test measuring orthogonal adhesion on the tens of micrometer to millimeter scale has been performed.
Refinement of the technique will hopefully bring about a universal mesoscale adhesion test. The orthogonal test technique measures the pure adhesive component of force, offering the opportunity for better standardization of adhesion testing. In contrast, lateral force measurements include a frictional force difficult to decouple from the adhesive force. The common peel test, while useful for an industrial standard for tapes, also introduces variables again difficult to decouple from adhesion, e.g. rate and the effect of the support medium.
The adhesive force was found to significantly depend on the maximum applied normal load (Fig. 6-7). This dependence is likely due to the increase in contact area between the two surfaces as a result of increased conformation, and deformation (plastic and elastic), of the adhesive to the spherical indenter surface. With hard flat substrates (e.g. silicon) clean loading and unloading curves were produced with no apparent adhesion. The photoresist surfaces demonstrated little adhesion (Fig. 6). In contrast, the organorod coated surfaces demonstrated much higher adhesion strengths (Fig. 6). Combining the organorods with the compliant pillar structures offered a significant increase in adhesion (Fig. 6), suggesting that the compliant structures aid in increasing the surface contact area. Initially it was anticipated that the pillar would allow the entire structure to rotate aiding in surface conformation.