Royal Society Publishing

Shearing of fibrillar adhesive microstructure: friction and shear-related changes in pull-off force

M Varenberg, S Gorb


To characterize the effect of shearing on function of fibrillar adhesive microstructure, friction and shear-related changes in pull-off force of a biomimetic polyvinylsiloxane mushroom-shaped fibrillar adhesive microstructure were studied. In contrast to a control flat surface, which exhibited pronounced stick–slip motion accompanied with high friction, the fibrillar microstructure demonstrated a stable and smooth sliding with a friction coefficient approximately four times lower. The structured contact also manifested zero pull-off force in a sheared state, while the flat surface exhibited highly scattered and unreliable pull-off force when affected by contact shearing. It appears that the fibrillar microstructure can be used in applications where a total attachment force should be generated in a binary on/off state and, most surprisingly, is suitable to stabilize and minimize elastomer friction.


1. Introduction

Natural systems hide a virtually endless potential of technological ideas for the development of new smart-performance surfaces and materials (Scherge & Gorb 2001; Gorb 2006), and detailed studies of nature's tribological solutions have become the focus of interdisciplinary scientific attention in recent years. The principles of contact splitting characterizing various natural attachment systems have been intensively discussed in several publications (Arzt et al. 2003; Persson 2003; Persson & Gorb 2003; Chung & Chaudhury 2005; Gao et al. 2005; Varenberg et al. 2006a) and, following extensive studies of biological fibrillar systems, few attempts have been made to produce artificial patterned adhesives (Geim et al. 2003; Ghatak et al. 2004; Glassmaker et al. 2004; Majidi et al. 2004; Peressadko & Gorb 2004; Crosby et al. 2005; Northen & Turner 2005; Yurdumakan et al. 2005). However, the reported surfaces exhibited limited life cycle and overall gain in adhesion.

Based on the analysis of the functional morphology of the tarsal hairs found in numerous species of beetles from the family Chrysomelidae, an advanced biomimetic mushroom-shaped fibrillar adhesive microstructure has been recently fabricated (Gorb et al. 2007). Initial testing revealed that the adhesive properties and the tolerance to contamination of this microstructure were much more effective than those of a control flat surface made of the same soft synthetic polymer. However, the tests conducted did not examine the effect of shearing, which may become significant in using this material under field conditions. In addition, despite the well-known fact that the animals possessing fibrillar attachment systems frequently use shear motion to either enhance or eliminate adhesion (Autumn et al. 2000; Gorb 2001; Niederegger & Gorb 2003; Autumn et al. 2006), the sliding behaviour of artificial fibrillar surfaces has only recently started to be investigated (Daltorio et al. 2005; Majidi et al. 2006). In light of the above, the purpose of the present work was to study friction and shear-related changes in pull-off force resulting from the shearing of biomimetic mushroom-shaped fibrillar adhesive microstructure.

2. Experimental details

The tests were performed on a home-made microtribometer that consists of motorized translation stages M-011 (Physik Instrumente GmbH, Karlsruhe, Germany) and fixed tensometric force transducers FORT-100 (World Precision Instruments Inc., Sarasota, Florida), whose bending in normal and lateral directions is used to determine the applied and friction forces with a resolution of 0.1 mN (figure 1). The force transducers were calibrated with known precise weights immediately prior to testing. To guarantee full contact during measurements in a flat-on-flat contact scheme essential in surface texture testing, a passive self-aligning system of specimen holders was used (Varenberg et al. 2006b). Specimens tested were flat and structured (figure 2; Gottlieb Binder GmbH, Holzgerlingen, Germany) polyvinylsiloxane (PVS) discs of 2 mm in diameter and 1 mm in height mounted on the force transducer. A glass slide of 18×7×0.2 mm3 size fixed on the translation stage was used as a substrate. Structured surface consisted of uniformly distributed mushroom-shaped pillars of approximately 100 μm in height bearing terminal contact plates of approximately 40 μm in diameter (Gorb et al. 2007). The area density of the terminal contact plates was approximately 40%. The roughness average (Ra) of the glass and PVS surface was approximately 1 and 85 nm, respectively. Before the experiments, the specimens were washed with deionized water and liquid soap and then dried in flowing nitrogen. The temperature and relative humidity in the laboratory were 24°C and 41%, respectively.

Figure 1

Schematic of the microtribometer.

Figure 2

Contact surface of (a) flat and (b) structured PVS specimens.

Each test started by bringing the PVS specimens in contact with the glass substrate. After applying a normal load of 40, 80, 120 or 160 mN, the translation stage was moved in the lateral direction at a velocity of 100 μm s−1 for distances ranging from 25 μm to 7 mm, while the tangential force resisting the specimen motion was measured. Immediately before completing the lateral motion, the contact area was imaged with a monochrome digital camera M4+CL (JAI Co., Yokohama, Japan) enhanced by high-magnification optics Zoom-12X (Navitar, Inc., Rochester, New York) in order to examine the contact behaviour in shearing. Finally, the pull-off force affected by shearing of the contact zone was measured while withdrawing the translation stage in a vertical direction at a velocity of 100 μm s−1.

The output signals from the microtribometer were acquired with a multifunctional data acquisition board PCI-6251 (National Instruments Co., Austin, Texas). The translation stages were controlled in a closed-loop manner with a stepper/servo motion controller PCI-7344 (National Instruments Co.). The contact images were obtained with an image acquisition board PCI-1428 (National Instruments Co.). To provide online monitoring of the contact response, all data were processed with a LabVIEW software package.

3. Results and discussion

Figure 3 shows representative friction curves obtained as a function of lateral stage displacement under an applied load of 40 mN. The difference in sliding behaviour of flat and structured surfaces becomes obvious. The flat surface exhibited pronounced stick–slip motion (figure 3a), which was rather similar to specimen jumps and associated with complete detachment of the PVS surface from the glass substrate. These jumps presumably arise from the self-aligning ability of the contacting surfaces and considerably differ from the ‘waves of detachment’ (Schallamach 1971), characterizing relative motion of fixed elastomer surfaces. This behaviour was clearly identified by optical inspection of the contact zone during the tests, as destructive interference of reflected white light in the glass–PVS interface resulted in visualization of the real contact area. Inserts a1–a4, where the dark grey zone corresponds to the real contact and the zones coloured from light grey to white are non-contact regions, demonstrate changes in contact area during one gross stick–slip event. Insert a1 presents an initial contact. No relative movement of the contacting surfaces is observed between the specimen jumps, and growing elastic stresses during the stick phase of motion result in reduction of a real contact area only (inserts a2, a3). After each jump, the surface comes in contact at different stress conditions that lead to differences in appearance of the real contact area (inserts a3, a4).

Figure 3

Friction force of (a) flat and (b) structured specimens measured as a function of lateral stage displacement under an applied load of 40 mN. Inserts a1, a2, a3 and a4 are images of flat contact area made after lateral displacements of 60, 460, 620 and 720 μm, respectively. Inserts b1, b2, b3 and b4 are images of structured contact area made after lateral displacements of 60, 200, 350 and 820 μm, respectively. DP, dust particles; FC, flat surface in contact; FO, flat surface out of contact; TC (dark grey spots), terminal plates in contact; PC (grey spots), bent pillars in contact; arrows in the direction of specimen motion.

In contrast to the flat surface, the structured one demonstrated a stable and smooth sliding (figure 3b) characterized by a significantly lower lateral force needed to initiate a relative motion between the contacting surfaces, which is in agreement with the preliminary results reported earlier (Daltorio et al. 2005). The microstructured contact behaviour during sliding is depicted in inserts b1–b4, where the first three images demonstrate an initial transient period and the last one presents the steady-state motion. It is shown that the flat plates terminating the structured surface gradually lose their contact with the glass substrate until the real contact area becomes much smaller and the whole contact starts sliding. A growing tangential load produced by the contact shearing is accommodated by bending of individual pillars, which bear the normal load on their edges when the terminal plates tend to come out of the contact (figure 4). Interestingly, every pillar is independent of its neighbours and, despite individual pillars possibly experiencing sliding irregularities, their collective motion results in a smooth sliding of the entire surface due to a random distribution and small amplitude of single stick–slip events. Through a significant decrease in real contact area (figure 3b, inserts 1–4; figure 4), which determines the actual friction force (Parker & Hatch 1950), an individual pillar bending is also believed to contribute to a reduction in the static friction observed with the structured surface. This is the opposite of the microfibre arrays constructed from a stiff polypropylene, where the microstructure deformation leads to an increase in real contact area and the fibre arrays show more than an order of magnitude increase in friction compared with the bulk material (Majidi et al. 2006).

Figure 4

Sheared PVS microstructure in contact with a flat glass substrate. Black areas are the real contact zones formed by both terminal plates and stalks of bent pillars.

Figure 5 depicts static friction force of flat and structured specimens measured as a function of applied load. The friction force is linear with respect to the applied load in both the cases, while the flat surface demonstrates much higher friction. It is worth noting that the friction force measured with the flat specimens is probably underestimated. The reason is that the self-aligning holders suspended on threads (Varenberg et al. 2006b) give to the specimens some freedom to move in normal direction and hence the specimens can jump before the contact is actually sheared. The friction coefficient μ, which is defined as a ratio of the friction force to the sum of the applied load and the pull-off force, is actually the slope of the friction–load line. It was fitted with the coefficient of determination R2 of 0.99 and found to be 1.34 and 0.37 for flat and structured surfaces, respectively. The pull-off force presented in this figure was determined before the friction tests and, according to the previous reports (Varenberg et al. 2006a,b), was found to be independent of maximum applied load in a loading cycle.

Figure 5

Static friction force of flat and structured specimens measured as a function of applied load.

Figure 6 shows pull-off force of flat and structured specimens affected by contact shearing under applied loads of 40, 80, 120 and 160 mN and measured as a function of lateral stage displacement. Shearing of both flat and structured surfaces initially led to reduction in the pull-off force, which approached zero at the lateral stage displacement of approximately 500 μm. With additional increase in lateral displacement, the pull-off force scattered between its unaltered value and zero on flat surfaces and remained zero on structured surfaces. Interestingly, despite presumably different physics in shearing of flat and structured surfaces, both the studied types of specimens lose an ability to resist pull-off forces after approximately the same critical lateral displacement. This, however, can be explained by the fact that deformation of the specimens and their holders accommodates a relatively large lateral displacement, while the difference in real interface displacement of flat and structured surfaces cannot be seen on this scale.

Figure 6

Pull-off force of flat and structured specimens affected by contact shearing under applied loads of 40, 80, 120 and 160 mN and measured as a function of lateral stage displacement.

A gradual loss of ability to resist pull-off forces with the application of a tangential load to the flat surface is connected to the stick–slip motion. No relative movement can be initiated in a large flat contact, while adhesive bonds between PVS and glass can withstand indefinitely a certain maximum tangential stress. Hence, the closer is the contact to the point at which the adhesive bonds are ruptured and the surface detachment occurs, the smaller is the additional pull-off force needed to withdraw the specimen from the contact. This is directly confirmed by figure 3a, where some stick–slip event can be seen at the lateral stage displacement of approximately 500 μm characteristic of a complete loss of ability to resist pull-off forces. The pull-off force measured at larger lateral displacements is scattered for the same reason, as after each detachment the surface jumped in contact at different tangential stress conditions that led to differences in withdrawal forces.

The mechanism of pull-off force reduction in the structured contact (figure 6) is most probably defined by pillar bending, which increases with the lateral stage displacement until attaining a certain level determined by a static friction force and remains constant after sliding inception. The degree of pillar bending is reflected in a gradual reduction of the contact zone (figure 3b, inserts 1–4) and storage of additional elastic energy working against adhesion. This directly affects the pull-off force, which is finally turned off at a definite lateral displacement. Interestingly, symmetric design of mushroom-shaped pillars leads to bending isotropy and does not allow controlled peeling of individual contact elements. Therefore, the pull-off force decreases independently of the direction of shearing. This is in contrast to asymmetric peeling-oriented spatula-shaped adhesive outgrowths found in attachment organs of different animals (Autumn et al. 2000; Gorb 2001; Gao et al. 2005), which control attachment forces by shearing the contact in different directions, thus changing the peeling angle of terminal contact elements (Niederegger & Gorb 2003; Autumn et al. 2006).

4. Conclusion

The tests performed show that from the viewpoint of both frictional and adhesive properties defined by contact shearing, the structured surface behaves in a much more predictable and simple way than the flat one, thus expanding the potential of this surface texturing. It appears that the fibrillar microstructure can also be used in applications where a total attachment force should be generated in a binary on/off state and, most surprisingly, is suitable to stabilize and minimize elastomer friction.


S.G. was supported in this work by the Federal Ministry of Education, Science and Technology, Germany (project InspiRat 01RI0633C).


    • Received December 20, 2006.
    • Accepted February 1, 2007.


View Abstract