Author’s Accepted Manuscript Sticky fingers: adhesive properties of human fingertips Marlene Spinner, Anke B. Wiechert, Stanislav N. Gorb www.elsevier.com/locate/jbiomech
PII: DOI: Reference:
S0021-9290(16)30061-6 http://dx.doi.org/10.1016/j.jbiomech.2016.01.033 BM7552
To appear in: Journal of Biomechanics Received date: 10 July 2015 Revised date: 18 January 2016 Accepted date: 28 January 2016 Cite this article as: Marlene Spinner, Anke B. Wiechert and Stanislav N. Gorb, Sticky fingers: adhesive properties of human fingertips, Journal of Biomechanics, http://dx.doi.org/10.1016/j.jbiomech.2016.01.033 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Sticky fingers: adhesive properties of human fingertips Marlene Spinner*, Anke B. Wiechert, Stanislav N. Gorb Zoological Institute, Kiel University, Am Botanischen Garten 9, 24118 Kiel, Germany * corresponding author email: [email protected] phone: +49 4318804506 fax: +49 431 8801389
Abstract Fingertip friction is a rather well studied subject. Although the phenomenon of finger stickiness is known as well, the pull-off force and the adhesive strength of human finger tips have never been previously quantified. For the first time, we provided here characterization of adhesive properties of human fingers under natural conditions. Human fingers can generate a maximum adhesive force of 15 mN on a smooth surface of epoxy resin. A weak correlation of the adhesive force and the normal force was found on all test surfaces. Up to 300 mN load, an increase of the normal force leads to an increase of the adhesive force. On rough surfaces, the adhesive strength is significantly reduced. Our data collected from untreated hands give also an impression of an enormous scattering of digital adhesion depending on a large set of inter-subject variability and time-dependent individual factors (skin texture, moisture level, perspiration). The wide inter- and intra-individual range of digital adhesion should be considered in developing of technical and medical products. . 1. Introduction The understanding of mechanical properties of human fingertips is of great importance for various scientific disciplines. The fingertip adhesion is a significant parameter in the development and improvement process of medical products and manual interfaces of technical devices. Although everybody knows the phenomenon of sticky fingertips occurring after pressing the finger against the smooth surface, up to now the perpendicular adhesive force of human fingertips has never been quantified by experimental data. Pailler-Mattéi and Zahouani (2006) and Pailler-Mattéi and coworkers (2007) analyzed the adhesive properties of human skin by indentation tests, but their measurements were carried out at the inner forearm, not on digital pads. Since microstructure, moisture content, and mechanical properties of different skin areas are very different, these published data cannot be directly 1
taken to understand fingertips adhesion. Among the large number of publications on frictional properties of human skin (reviewed in Tomlinson et al., 2007; Derler and Gerhardt, 2012; Van der Heide et al., 2013), experiments on human skin suggest that adhesion may be a strong factor contributing to friction forces (Wolfram, 1983; Adams et al., 2007; Kwiatkowska et al., 2009). During the sliding process, temporary junctions, caused by intermolecular adhesive forces, have to be ruptured, which potentially leads to stick-slip motion. However, also these studies provide no information on the range of adhesive forces occurring, when a finger tip detaches from a surface. The goal of this study was to evaluate the amount of adhesive strength of human fingers under natural conditions. We quantified the adhesive force of the index finger and the thumb on surfaces having same chemistry, but different texture profiles. Measurements made under a wide range of load conditions aided in drawing a comprehensive image of the controllability of adhesion by the grasping strength.
2. Material and Methods Adhesive forces of the volar pads of the tips (Phalanx distalis) of the left hand‘s index finger (Digitus secundus) and the thumb (Pollex) were measured in five female and five male test persons (24 - 32 years) under varying normal load ranging from 0 to 1N at T = 19.9 - 21.3°C and RH = 27.7 - 38.6% on four substrates: a smooth microscopy slide (Carl Roth GmbH & Co. KG, Karlsruhe, Germany), two polish papers of 1 and 12 µm grain size (FiberMet® Abrasive Discs, Buehler, Illinois, USA) and a sand paper P320 (Carbo Schröder, Hannover, Germany). In order to exclude the influence of materialspecific surface chemistry, resin replicas of these substrates were made. A polyvinylsiloxane (PVS) polymer negative (AFFINIS®, light body, ISO 4823, Type 3, low consistency, Coltène/Whaledent AG, Altstätten, Schwitzerland) of 2.56 cm² of each substrate was made and repeatedly filled with 1g of fluid epoxy resin with a mixture ratio of ERL = 10.0, D.E.R = 6.0, NSA = 26.0, and DMAE = 0.4 (see Spurr, 1969). The polymerised resin had a Young´s modulus of 7 GPa (Peisker and Gorb, 2010). The grain size of the test surfaces was determined from 60 grains of three scanning electron micrographs by using the software ImageJ (Version 1.43u, Wayne Rasband, National Institutes of Health, USA). Therefore, samples of all surfaces were sputter coated with a 10 nm layer of gold palladium (Leica SCD500, Leica Microsystems GmbH, Wetzlar, Germany) and subsequently examined in the SEM HITACHI S4800 (Hitachi High-Technologies Corporation, Tokyo, Japan). The sand paper P320 had a grain size of 55.3 ± 14.98 µm (manufacturer's information: 46.2 ± 1.5 µm). The polish paper of 1 µm and 12 µm had a grain size of 1.3 ± 0.54 and 10.8 ± 4.1 µm and an average roughness (Ra) of 0.38 ± 0.05 µm and 3.39 ± 1.02 µm and a root mean squared roughness (RMS) of 0.47 ± 0.06 µm and 4.16 ± 0.97 µm (data from Spinner et al., 2014). For the adhesion measurements a 100 g force sensor (World Precision Instruments, Inc, Sarasota, USA) connected with a BIOPAC System (System and software AcqKnowledge® Version 3.7.3, 2
BIOPAC Systems Inc, Santa Barbara, USA) was equipped with one of the test surfaces and tapped shortly in a vertical direction (down to the finger pad oriented upwards) with the underside of the apical digital pads and thereafter pulled off again. The test persons did not wash their hands at least one hour before the experiment. Continuous detection of forces, acting on the sensor, allowed to quantify the normal forces between the finger and the test surface, as well as to record the adhesive forces during the detachment process (Supplementary Fig. 1). The different test surfaces were pressed for 1.5 s against the finger with varying normal force for a time period of 60 s with as many repetitions as possible. The four test surfaces were changed in a random order so that each surface roughness was tested four times. For each test person, a new set of test surfaces was produced. The individual force measurements (= curves) were analyzed separately: The normal load was defined as the difference between the curve maximum and the offset defined by the noise level (Supplementary Fig. 1). Fingertip adhesion was in turn defined as difference between the curve minimum and the noise level (Supplementary Fig. 1). The level of noise was determined by calculation of the mean values of 200 data points (400 ms) before the sensor was pressed against the fingertip (Supplementary Fig. 1). The lowest value for the adhesive forces, taken into account in our subsequent analysis, depended on the mean value of noise and was about 2.8 mN. A total number of 3200 measurements was analyzed. The final data set included 320 curves per test person (resulting from four measurements with ten repetitions at the thumb and the index finger on each test surface).
3. Results
Our study demonstrates that human digital pads of both tested fingers, the index and the thumb, were adhesive on the whole tested range of surface roughness and at all levels of normal load (Fig. 1). The adhesive forces of the fingertips of the volunteers participating in this study ranged from 2.3 to 167.6 mN (Fig. 1). The strongest adhesion of 14 mN (median) was measured on smooth epoxy resin under a normal load of more than 300 mN. On the rough substrates of 1.3, 10.8, and 55.0 µm grain size, the maximum digital adhesion was considerably lower than that on the smooth surface (Figs. 2, 3). Analysis of pooled data of the index finger and thumb of all test persons showed that the digital adhesion was significantly different on all four tests surfaces under the different categories of normal load (low normal load = 0 - 300 mN: Kruskal-Wallis-test, χ2 = 115.182, df = 3, p = 0.000; medium normal load = 300 - 600 mN: Kruskal-Wallis test, χ2 = 68.084, df = 3, p = 0.000; high normal load = 600 - 1000 mN: Kruskal-Wallis test, χ2 = 43.479, df = 3, p = 0.000). While no significant difference was found among the three rough surfaces, adhesion on smooth epoxy was significantly higher than that on each of the other three surfaces within the whole range of tested load (0 - 1000 mN) (Pairwise Wilcoxon signed rank sum test, P ≤ 0.001) (Fig. 2). On all tested surfaces, we found also significant differences in adhesion among the ten different 100 mN load categories (Kruskal-Wallis test). In pairwise comparisons (Wilcoxon signed rank sum test) pooled data in the load categories lower than 3
300 mN were in most of the cases significantly different from the 100 mN load categories between 300 and 1000 mN, but also differed from each other (Fig. 3). Hence, the measured adhesive forces of human digital pads were significantly lower at low pressure than at medium and high pressure (>300 mN) and the adhesion increased with an increasing normal load within the loading regime of 0-300 mN. A positive correlation indicating a dependency of finger pad adhesion on compressive force in humans was also visible in the regression lines calculated by the LOWESS-method (locally weighted scatterplot smoothing, s = 0.25) (Fig. 1) and could be demonstrated by a Spearman rank correlation test (Supplementary Table). The load dependent increase of the apparent contact area was visualized with a glass plate that was pressed under varying normal load against the index finger (Fig. 4).
4. Discussion
Our study provides evidence for the presence of rather substantial adhesive forces on the apical digital pads in humans. The maximal adhesive force was revealed on the smooth test surface (168 mN), but even on the rough surfaces, we were able to detect some adhesion. We could therefore demonstrate that humans are able to pick up flat glass objects weighting up to 17.1 g with just one fingertip (Fig. 5). As another result, our measurements on the untreated skin of different individuals revealed a remarkably high deviation of the adhesive strength. Friction measurements on human skin of different degree of hydration (Derler and Gerhardt, 2012; Qian and Gao, 2006; Tomlinson et al., 2011; Pasumarty et al., 2011; Adams et al., 2013; Derler et al., 2015) or test subjects of different age and sex (Cua et al., 1990; 1995) under laboratory conditions, lead us already to expect a great influence of individual and external factors. Although for each test person an individual set of new test surfaces was used, sebum might have also remained on the test surfaces. The effect of fingerprints on contact mechanics has been recently demonstrated in a tribological study repeating Leonardo da Vinci´s friction experiments (Pitenis et al., 2014). Our measurements showed that in real life the combined effect of these variables causes high differences in the human adhesive strength that have to be seriously taken into account when tuning haptic properties of surfaces in technical products. In general, our study revealed a significant positive correlation between the adhesive forces and the preceding normal forces, irrespective of the surface profile. Our data clearly showed that up to a certain threshold of applied load (here 300 mN), the adhesive force of human digital pads increases with an increasing normal load. Also the surface roughness had a significant influence on the forces that have to be generated to detach the finger from the surface. Both phenomena, the load and roughness dependency of the human adhesive strength can be explained by anatomical factors influencing the geometry of the contact. Adhesion can only be generated, if the external layer of the finger, the stratum corneum, is in an intimate contact with the test surface, or if the distance between the skin and test surface is so small that a third body (fluid substances, such as water, sebum or sweat) 4
can generate capillary/viscous forces or even function as glue. The positive correlation between the normal load and the apparent (macroscopic) contact area of human fingertips has been investigated in numerous studies reviewed in van Kuilenburg et al., 2013 (Kinoshita et al., 1997; Maeno et al., 1998; Childs and Henson, 2007; Soneda and Nakano, 2008; Warman and Ennos, 2009; Derler et al., 2009; Tomimoto, 2011; André et al., 2011). The real contact area corresponds to the overall area of ridges (see Supplementary Fig. 2) on the fingertip getting in contact with the surface. As for the apparent contact area, experimental studies found a load dependent increase of the contact area (Tomlinson, 2009; Soneda and Nakano, 2010). It is very likely that load-dependent increase of adhesion is mainly caused by an increase of the amount of digital ridges getting in contact with the test surfaces. Applying the apparent (99.4 mm²) and the real area of contact (29.9 mm²) of a finger tip measured at 1000 mN (Soneda and Nakano, 2010), an adhesive strength of about 150 and 500 Pa can be calculated on the smooth resin surface of this study. Compared with the adhesive systems of other vertebrates such as the setal foot pads of the Tokay gecko (Gekko gecko) (Autumn et al., 2002) this value is quite low. It is obvious that in contrast to their primate relatives the capability to walk on vertical locomotion was not the major selective factor in the evolution of human hands. However, adhesion might also be an excellent preadaptation for the ability to grasp, manipulate, modify and control objects (Bishop, 1962). During the precision grip, where an object is held between thumb tip and index fingertip (Bishop, 1962), an increase of the pressing forces from 0 to 300 mN leads to an adhesion increase. The normal load can thus be used as controlling parameter of adhesion. Variation of the time span of the surface contact could be considered as further parameter of control. In addition to friction, the adhesion forces might also represent additional information about the surface properties detectable by finger tactile sensors (van Kuilenburg et al., 2013).
Ethics approval
This study was permitted in advance by the ethics commission of the Medical Faculty of Kiel University (reference number D 516/13).
Conflicts of interests
Competing interests: None declared.
Acknowledgements
5
The project CP 1208 was financially supported by a grant of the Cluster of Excellence 80 ‘The Future Ocean’ to MS. ‘The Future Ocean’ is funded within the framework of the Excellence Initiative by the Deutsche Forschungsgemeinschaft (DFG) on behalf of the German federal and state governments.
Authorship
All authors contributed equally to the experimental design and the interpretation of the data.. A. B. W. made the measurements and statistical analysis and prepared the figures and tables. M. S. wrote the manuscript.
References
Adams, M., Briscoe, B., Johnson, S., 2007. Friction and lubrication of human skin. Tribology Letters 26, 239-253.
Adams, M. J., Johnson, S. A., Lefèvre, P., Lévesque, V., Hayward, V., André, T., Thonnard, J. L., 2013. Finger pad friction and its role in grip and touch. Journal of The Royal Society Interface 10, 20120467.
André, T., Lévesque, V., Hayward, V., Lefèvre, P., Thonnard, J. L., 2011. Effect of skin hydration on the dynamics of fingertip gripping contact. Journal of The Royal Society Interface 8, 1574-1583.
Autumn, K., Sitti, M., Liang, Y. A., Peattie, A. M., Hansen, W. R., Sponberg, S., Kenny, T. W., Fearing, R., Israelachvili, J. N., Full, R. J., 2002. Evidence for van der Waals adhesion in gecko setae. Proceedings of the National Academy of Sciences 99, 12252-12256.
Bishop, A., 1962. Control of the Hand in Lower Primates. Annals of the New York Academy of Sciences 102, 316-337.
Childs, T. H. C., Henson, B., 2007. Human tactile perception of screen-printed surfaces: selfreport and contact mechanics experiments. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 221, 427-441.
Cua, A. B., Wilhelm, K. P., Maibach, H. I., 1990. Frictional properties of human skin: relation to age, sex and anatomical region, stratum corneum hydration and transepidermal water loss. British Journal of Dermatology 123, 473-479. 6
Cua, A. B., Wilhelm, K. P., Maibach, H. I., 1995. Skin surface lipid and skin friction: relation to age, sex and anatomical region. Skin Pharmacology and Physiology 8, 246-251.
Derler, S., Gerhardt, L. C., 2012. Tribology of skin: review and analysis of experimental results for the friction coefficient of human skin. Tribology Letters 45, 1-27.
Derler, S., Gerhardt, L. C., Lenz, A., Bertaux, E., Hadad, M., 2009. Friction of human skin against smooth and rough glass as a function of the contact pressure. Tribology International 42, 1565-1574.
Derler, S., Rossi, R. M., Rotaru, G. M., 2015. Understanding the variation of friction coefficients of human skin as a function of skin hydration and interfacial water films. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 229, 285-293.
Kinoshita, H., Bäckström, L., Flanagan, J. R., Johansson, R.S., 1997. Tangential torque effects on the control of grip forces when holding objects with a precision grip. Journal of Neurophysiology 78, 1619-1630.
Kwiatkowska, M., Franklin, S. E., Hendriks, C. P., Kwiatkowski, K., 2009. Friction and deformation behaviour of human skin. Wear 267, 1264-1273.
Maeno, T., Kobayashi, K., Yamazaki, N., 1998. Relationship between the structure of human finger tissue and the location of tactile receptors. JSME International Journal Series C 41, 94100.
Pailler-Mattéi, C., Zahouani, H., 2006. Analysis of adhesive behaviour of human skin in vivo by an indentation test. Tribology International 39, 12-21.
Pailler-Mattéi, C., Pavan, S., Vargiolu, R., Pirot, F., Falson, F., Zahouani, H., 2007. Contribution of stratum corneum in determining bio-tribological properties of the human skin. Wear 263, 1038-1043.
Pasumarty, S. M., Johnson, S. A., Watson, S. A., Adams, M. J., 2011. Friction of the human finger pad: influence of moisture, occlusion and velocity. Tribology Letters 44, 117-137.
7
Peisker, H., Gorb, S. N., 2010. Always on the bright side of life: anti-adhesive properties of insect ommatidia grating. Journal of Experimental Biolology 213, 3457-3462. Pitenis, A. A., Dowson, D., Sawyer, W. G. 2014. Leonardo da Vinci’s Friction Experiments: An Old Story Acknowledged and Repeated. Tribology Letters 56, 509-515.
Qian, J., Gao, H., 2006. Scaling effects of wet adhesion in biological attachment systems. Acta Biomaterialia 2, 51-58.
Soneda, T., Nakano, K, 2008. Investigation of the vibration detectability of human fingers supported by the observation of contact zones. In: Proceedings of the 2nd international conference on advanced tribology, Singapore iCAT288.
Soneda, T., Nakano, K., 2010. Investigation of vibrotactile sensation of human fingerpads by observation of contact zones. Tribology International 43, 210-217.
Spinner M., Westhoff G., Gorb S. N., 2014. Subdigital setae of chameleon feet: Frictionenhancing microstructures for a wide range of substrate roughness. Scientific Reports 4:5481.
Spurr, A. R., 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. Journal of Ultrastructure Research 26, 31-43.
Tomimoto, M., 2011. The frictional pattern of tactile sensations in anthropomorphic fingertip. Tribology International 44, 1340-1347.
Tomlinson, S. E., 2009. Understanding the friction between human fingers and contacting surfaces. PhD thesis, The University of Sheffield.
Tomlinson, S. E., Lewis, R., Carré, M. J., 2007. Review of the frictional properties of finger– object contact when gripping. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 221, 841-850.
Tomlinson, S. E., Lewis, R., Liu, X., Texier, C., Carré, M. J., 2011. Understanding the friction mechanisms between the human finger and flat contacting surfaces in moist conditions. Tribology Letters 41, 283-294.
8
Van der Heide, E., Zeng, X., Masen, M. A., 2013. Skin tribology: science friction? Friction 1, 130-142.
van Kuilenburg, J., Masen, M. A., van der Heide, E., 2013. A review of fingerpad contact mechanics and friction and how this affects tactile perception. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 0, 1-16.
Warman, P. H., Ennos, A. R., 2009. Fingerprints are unlikely to increase the friction of primate fingerpads. Journal of Experimental Biology 212, 2016-2022.
Wolfram, L. J., 1983. Friction of skin. Journal of the Society of Cosmetic Chemists 34, 465476.
Fig. 1. Adhesive forces of human fingers. The adhesive forces of the tips of index finger (blue circles) and the thumb (red crosses) under different normal load ranging from 0 to 1000 mN measured on epoxy resin replica of a smooth glass surface (a), and surfaces of a grain size of 1.3 µm (b), 10.8 µm (c), and 55.0 µm (d). Scanning electron micrographs of the test surface are presented on the top left 9
side of each graph. The scale bar - 20 µm. The regression lines for measurements of the index finger (black line) and the thumb (dashed line) were calculated by the LOWESS-method.
Fig. 2. Adhesive forces of the human index finger and the thumb (pooled) on resin replicas of different surface roughness under different load conditions: a) 0-300 mN, b) 300-600 mN c) 600-1000 mN. Boxes and horizontal lines represent the upper and lower quartiles and the median values, respectively. Whiskers show the maximum and minimum values. Bars and asterisks indicate significant difference.
10
Fig. 3. The effect of the normal load. Results of pairwise comparisons (pairwise Wilcoxon signed rank sum test) between the ten 100 mN pressure categories for each test surface: a) smooth surface, b) 1.3 µm grain size, c), 10.8 µm grain size, and d) 55.0 µm grain size. Significance levels are indicated by asterisks (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001).
Fig. 4. Relation between the normal load and the apparent contact area. The increase and decrease of normal load (a) and apparent contact area (b) during the experiment. An increase of the normal load leads to an increase of the apparent contact area.
Fig. 5. A fingertip in contact with a glass surface. A glass plate can be hold by the adhesive forces of the tip of a human index finger. Supplementary
Supplementary Fig.1. An example of typical recorded force curve. 1. Test surface is pressing against the finger tip. The sensor detects an increase of normal load force. 2-3. Test surface is removing from the finger tip and the detected force decreases. 3. The detected force is falling under zero level, indicating the presence of the finger tip adhesion to the test surface. 4. Contact rupture. The mean 11
value of noise (N0) was calculated over 400 ms before the measurement. The normal force corresponds to the difference between the maximum force curve and N0, whereas the adhesive force corresponds to the difference between the minimum of the force curve and N0. Supplementary Fig. 2. Fingerprints and surface morphology of the human finger tip. a) and b) Film of stratum corneum cells, sweat, and sebum (fingerprint) released on a glass plate in the stereo microscope (Leica M205 A, Leica Microsystems, Wetzlar, Germany) under the coaxial illumination. c) and d) SEM image (HITACHI S800, Hitachi High-Technologies Corporation, Tokyo, Japan).of the two-step positive mould of the living human finger. Note droplets of the fluid (SG) coming out of the pores of sweat glands. RG, ridges of the finger skin.
12
PII: DOI: Reference:
S0021-9290(16)30061-6 http://dx.doi.org/10.1016/j.jbiomech.2016.01.033 BM7552
To appear in: Journal of Biomechanics Received date: 10 July 2015 Revised date: 18 January 2016 Accepted date: 28 January 2016 Cite this article as: Marlene Spinner, Anke B. Wiechert and Stanislav N. Gorb, Sticky fingers: adhesive properties of human fingertips, Journal of Biomechanics, http://dx.doi.org/10.1016/j.jbiomech.2016.01.033 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Sticky fingers: adhesive properties of human fingertips Marlene Spinner*, Anke B. Wiechert, Stanislav N. Gorb Zoological Institute, Kiel University, Am Botanischen Garten 9, 24118 Kiel, Germany * corresponding author email: [email protected] phone: +49 4318804506 fax: +49 431 8801389
Abstract Fingertip friction is a rather well studied subject. Although the phenomenon of finger stickiness is known as well, the pull-off force and the adhesive strength of human finger tips have never been previously quantified. For the first time, we provided here characterization of adhesive properties of human fingers under natural conditions. Human fingers can generate a maximum adhesive force of 15 mN on a smooth surface of epoxy resin. A weak correlation of the adhesive force and the normal force was found on all test surfaces. Up to 300 mN load, an increase of the normal force leads to an increase of the adhesive force. On rough surfaces, the adhesive strength is significantly reduced. Our data collected from untreated hands give also an impression of an enormous scattering of digital adhesion depending on a large set of inter-subject variability and time-dependent individual factors (skin texture, moisture level, perspiration). The wide inter- and intra-individual range of digital adhesion should be considered in developing of technical and medical products. . 1. Introduction The understanding of mechanical properties of human fingertips is of great importance for various scientific disciplines. The fingertip adhesion is a significant parameter in the development and improvement process of medical products and manual interfaces of technical devices. Although everybody knows the phenomenon of sticky fingertips occurring after pressing the finger against the smooth surface, up to now the perpendicular adhesive force of human fingertips has never been quantified by experimental data. Pailler-Mattéi and Zahouani (2006) and Pailler-Mattéi and coworkers (2007) analyzed the adhesive properties of human skin by indentation tests, but their measurements were carried out at the inner forearm, not on digital pads. Since microstructure, moisture content, and mechanical properties of different skin areas are very different, these published data cannot be directly 1
taken to understand fingertips adhesion. Among the large number of publications on frictional properties of human skin (reviewed in Tomlinson et al., 2007; Derler and Gerhardt, 2012; Van der Heide et al., 2013), experiments on human skin suggest that adhesion may be a strong factor contributing to friction forces (Wolfram, 1983; Adams et al., 2007; Kwiatkowska et al., 2009). During the sliding process, temporary junctions, caused by intermolecular adhesive forces, have to be ruptured, which potentially leads to stick-slip motion. However, also these studies provide no information on the range of adhesive forces occurring, when a finger tip detaches from a surface. The goal of this study was to evaluate the amount of adhesive strength of human fingers under natural conditions. We quantified the adhesive force of the index finger and the thumb on surfaces having same chemistry, but different texture profiles. Measurements made under a wide range of load conditions aided in drawing a comprehensive image of the controllability of adhesion by the grasping strength.
2. Material and Methods Adhesive forces of the volar pads of the tips (Phalanx distalis) of the left hand‘s index finger (Digitus secundus) and the thumb (Pollex) were measured in five female and five male test persons (24 - 32 years) under varying normal load ranging from 0 to 1N at T = 19.9 - 21.3°C and RH = 27.7 - 38.6% on four substrates: a smooth microscopy slide (Carl Roth GmbH & Co. KG, Karlsruhe, Germany), two polish papers of 1 and 12 µm grain size (FiberMet® Abrasive Discs, Buehler, Illinois, USA) and a sand paper P320 (Carbo Schröder, Hannover, Germany). In order to exclude the influence of materialspecific surface chemistry, resin replicas of these substrates were made. A polyvinylsiloxane (PVS) polymer negative (AFFINIS®, light body, ISO 4823, Type 3, low consistency, Coltène/Whaledent AG, Altstätten, Schwitzerland) of 2.56 cm² of each substrate was made and repeatedly filled with 1g of fluid epoxy resin with a mixture ratio of ERL = 10.0, D.E.R = 6.0, NSA = 26.0, and DMAE = 0.4 (see Spurr, 1969). The polymerised resin had a Young´s modulus of 7 GPa (Peisker and Gorb, 2010). The grain size of the test surfaces was determined from 60 grains of three scanning electron micrographs by using the software ImageJ (Version 1.43u, Wayne Rasband, National Institutes of Health, USA). Therefore, samples of all surfaces were sputter coated with a 10 nm layer of gold palladium (Leica SCD500, Leica Microsystems GmbH, Wetzlar, Germany) and subsequently examined in the SEM HITACHI S4800 (Hitachi High-Technologies Corporation, Tokyo, Japan). The sand paper P320 had a grain size of 55.3 ± 14.98 µm (manufacturer's information: 46.2 ± 1.5 µm). The polish paper of 1 µm and 12 µm had a grain size of 1.3 ± 0.54 and 10.8 ± 4.1 µm and an average roughness (Ra) of 0.38 ± 0.05 µm and 3.39 ± 1.02 µm and a root mean squared roughness (RMS) of 0.47 ± 0.06 µm and 4.16 ± 0.97 µm (data from Spinner et al., 2014). For the adhesion measurements a 100 g force sensor (World Precision Instruments, Inc, Sarasota, USA) connected with a BIOPAC System (System and software AcqKnowledge® Version 3.7.3, 2
BIOPAC Systems Inc, Santa Barbara, USA) was equipped with one of the test surfaces and tapped shortly in a vertical direction (down to the finger pad oriented upwards) with the underside of the apical digital pads and thereafter pulled off again. The test persons did not wash their hands at least one hour before the experiment. Continuous detection of forces, acting on the sensor, allowed to quantify the normal forces between the finger and the test surface, as well as to record the adhesive forces during the detachment process (Supplementary Fig. 1). The different test surfaces were pressed for 1.5 s against the finger with varying normal force for a time period of 60 s with as many repetitions as possible. The four test surfaces were changed in a random order so that each surface roughness was tested four times. For each test person, a new set of test surfaces was produced. The individual force measurements (= curves) were analyzed separately: The normal load was defined as the difference between the curve maximum and the offset defined by the noise level (Supplementary Fig. 1). Fingertip adhesion was in turn defined as difference between the curve minimum and the noise level (Supplementary Fig. 1). The level of noise was determined by calculation of the mean values of 200 data points (400 ms) before the sensor was pressed against the fingertip (Supplementary Fig. 1). The lowest value for the adhesive forces, taken into account in our subsequent analysis, depended on the mean value of noise and was about 2.8 mN. A total number of 3200 measurements was analyzed. The final data set included 320 curves per test person (resulting from four measurements with ten repetitions at the thumb and the index finger on each test surface).
3. Results
Our study demonstrates that human digital pads of both tested fingers, the index and the thumb, were adhesive on the whole tested range of surface roughness and at all levels of normal load (Fig. 1). The adhesive forces of the fingertips of the volunteers participating in this study ranged from 2.3 to 167.6 mN (Fig. 1). The strongest adhesion of 14 mN (median) was measured on smooth epoxy resin under a normal load of more than 300 mN. On the rough substrates of 1.3, 10.8, and 55.0 µm grain size, the maximum digital adhesion was considerably lower than that on the smooth surface (Figs. 2, 3). Analysis of pooled data of the index finger and thumb of all test persons showed that the digital adhesion was significantly different on all four tests surfaces under the different categories of normal load (low normal load = 0 - 300 mN: Kruskal-Wallis-test, χ2 = 115.182, df = 3, p = 0.000; medium normal load = 300 - 600 mN: Kruskal-Wallis test, χ2 = 68.084, df = 3, p = 0.000; high normal load = 600 - 1000 mN: Kruskal-Wallis test, χ2 = 43.479, df = 3, p = 0.000). While no significant difference was found among the three rough surfaces, adhesion on smooth epoxy was significantly higher than that on each of the other three surfaces within the whole range of tested load (0 - 1000 mN) (Pairwise Wilcoxon signed rank sum test, P ≤ 0.001) (Fig. 2). On all tested surfaces, we found also significant differences in adhesion among the ten different 100 mN load categories (Kruskal-Wallis test). In pairwise comparisons (Wilcoxon signed rank sum test) pooled data in the load categories lower than 3
300 mN were in most of the cases significantly different from the 100 mN load categories between 300 and 1000 mN, but also differed from each other (Fig. 3). Hence, the measured adhesive forces of human digital pads were significantly lower at low pressure than at medium and high pressure (>300 mN) and the adhesion increased with an increasing normal load within the loading regime of 0-300 mN. A positive correlation indicating a dependency of finger pad adhesion on compressive force in humans was also visible in the regression lines calculated by the LOWESS-method (locally weighted scatterplot smoothing, s = 0.25) (Fig. 1) and could be demonstrated by a Spearman rank correlation test (Supplementary Table). The load dependent increase of the apparent contact area was visualized with a glass plate that was pressed under varying normal load against the index finger (Fig. 4).
4. Discussion
Our study provides evidence for the presence of rather substantial adhesive forces on the apical digital pads in humans. The maximal adhesive force was revealed on the smooth test surface (168 mN), but even on the rough surfaces, we were able to detect some adhesion. We could therefore demonstrate that humans are able to pick up flat glass objects weighting up to 17.1 g with just one fingertip (Fig. 5). As another result, our measurements on the untreated skin of different individuals revealed a remarkably high deviation of the adhesive strength. Friction measurements on human skin of different degree of hydration (Derler and Gerhardt, 2012; Qian and Gao, 2006; Tomlinson et al., 2011; Pasumarty et al., 2011; Adams et al., 2013; Derler et al., 2015) or test subjects of different age and sex (Cua et al., 1990; 1995) under laboratory conditions, lead us already to expect a great influence of individual and external factors. Although for each test person an individual set of new test surfaces was used, sebum might have also remained on the test surfaces. The effect of fingerprints on contact mechanics has been recently demonstrated in a tribological study repeating Leonardo da Vinci´s friction experiments (Pitenis et al., 2014). Our measurements showed that in real life the combined effect of these variables causes high differences in the human adhesive strength that have to be seriously taken into account when tuning haptic properties of surfaces in technical products. In general, our study revealed a significant positive correlation between the adhesive forces and the preceding normal forces, irrespective of the surface profile. Our data clearly showed that up to a certain threshold of applied load (here 300 mN), the adhesive force of human digital pads increases with an increasing normal load. Also the surface roughness had a significant influence on the forces that have to be generated to detach the finger from the surface. Both phenomena, the load and roughness dependency of the human adhesive strength can be explained by anatomical factors influencing the geometry of the contact. Adhesion can only be generated, if the external layer of the finger, the stratum corneum, is in an intimate contact with the test surface, or if the distance between the skin and test surface is so small that a third body (fluid substances, such as water, sebum or sweat) 4
can generate capillary/viscous forces or even function as glue. The positive correlation between the normal load and the apparent (macroscopic) contact area of human fingertips has been investigated in numerous studies reviewed in van Kuilenburg et al., 2013 (Kinoshita et al., 1997; Maeno et al., 1998; Childs and Henson, 2007; Soneda and Nakano, 2008; Warman and Ennos, 2009; Derler et al., 2009; Tomimoto, 2011; André et al., 2011). The real contact area corresponds to the overall area of ridges (see Supplementary Fig. 2) on the fingertip getting in contact with the surface. As for the apparent contact area, experimental studies found a load dependent increase of the contact area (Tomlinson, 2009; Soneda and Nakano, 2010). It is very likely that load-dependent increase of adhesion is mainly caused by an increase of the amount of digital ridges getting in contact with the test surfaces. Applying the apparent (99.4 mm²) and the real area of contact (29.9 mm²) of a finger tip measured at 1000 mN (Soneda and Nakano, 2010), an adhesive strength of about 150 and 500 Pa can be calculated on the smooth resin surface of this study. Compared with the adhesive systems of other vertebrates such as the setal foot pads of the Tokay gecko (Gekko gecko) (Autumn et al., 2002) this value is quite low. It is obvious that in contrast to their primate relatives the capability to walk on vertical locomotion was not the major selective factor in the evolution of human hands. However, adhesion might also be an excellent preadaptation for the ability to grasp, manipulate, modify and control objects (Bishop, 1962). During the precision grip, where an object is held between thumb tip and index fingertip (Bishop, 1962), an increase of the pressing forces from 0 to 300 mN leads to an adhesion increase. The normal load can thus be used as controlling parameter of adhesion. Variation of the time span of the surface contact could be considered as further parameter of control. In addition to friction, the adhesion forces might also represent additional information about the surface properties detectable by finger tactile sensors (van Kuilenburg et al., 2013).
Ethics approval
This study was permitted in advance by the ethics commission of the Medical Faculty of Kiel University (reference number D 516/13).
Conflicts of interests
Competing interests: None declared.
Acknowledgements
5
The project CP 1208 was financially supported by a grant of the Cluster of Excellence 80 ‘The Future Ocean’ to MS. ‘The Future Ocean’ is funded within the framework of the Excellence Initiative by the Deutsche Forschungsgemeinschaft (DFG) on behalf of the German federal and state governments.
Authorship
All authors contributed equally to the experimental design and the interpretation of the data.. A. B. W. made the measurements and statistical analysis and prepared the figures and tables. M. S. wrote the manuscript.
References
Adams, M., Briscoe, B., Johnson, S., 2007. Friction and lubrication of human skin. Tribology Letters 26, 239-253.
Adams, M. J., Johnson, S. A., Lefèvre, P., Lévesque, V., Hayward, V., André, T., Thonnard, J. L., 2013. Finger pad friction and its role in grip and touch. Journal of The Royal Society Interface 10, 20120467.
André, T., Lévesque, V., Hayward, V., Lefèvre, P., Thonnard, J. L., 2011. Effect of skin hydration on the dynamics of fingertip gripping contact. Journal of The Royal Society Interface 8, 1574-1583.
Autumn, K., Sitti, M., Liang, Y. A., Peattie, A. M., Hansen, W. R., Sponberg, S., Kenny, T. W., Fearing, R., Israelachvili, J. N., Full, R. J., 2002. Evidence for van der Waals adhesion in gecko setae. Proceedings of the National Academy of Sciences 99, 12252-12256.
Bishop, A., 1962. Control of the Hand in Lower Primates. Annals of the New York Academy of Sciences 102, 316-337.
Childs, T. H. C., Henson, B., 2007. Human tactile perception of screen-printed surfaces: selfreport and contact mechanics experiments. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 221, 427-441.
Cua, A. B., Wilhelm, K. P., Maibach, H. I., 1990. Frictional properties of human skin: relation to age, sex and anatomical region, stratum corneum hydration and transepidermal water loss. British Journal of Dermatology 123, 473-479. 6
Cua, A. B., Wilhelm, K. P., Maibach, H. I., 1995. Skin surface lipid and skin friction: relation to age, sex and anatomical region. Skin Pharmacology and Physiology 8, 246-251.
Derler, S., Gerhardt, L. C., 2012. Tribology of skin: review and analysis of experimental results for the friction coefficient of human skin. Tribology Letters 45, 1-27.
Derler, S., Gerhardt, L. C., Lenz, A., Bertaux, E., Hadad, M., 2009. Friction of human skin against smooth and rough glass as a function of the contact pressure. Tribology International 42, 1565-1574.
Derler, S., Rossi, R. M., Rotaru, G. M., 2015. Understanding the variation of friction coefficients of human skin as a function of skin hydration and interfacial water films. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 229, 285-293.
Kinoshita, H., Bäckström, L., Flanagan, J. R., Johansson, R.S., 1997. Tangential torque effects on the control of grip forces when holding objects with a precision grip. Journal of Neurophysiology 78, 1619-1630.
Kwiatkowska, M., Franklin, S. E., Hendriks, C. P., Kwiatkowski, K., 2009. Friction and deformation behaviour of human skin. Wear 267, 1264-1273.
Maeno, T., Kobayashi, K., Yamazaki, N., 1998. Relationship between the structure of human finger tissue and the location of tactile receptors. JSME International Journal Series C 41, 94100.
Pailler-Mattéi, C., Zahouani, H., 2006. Analysis of adhesive behaviour of human skin in vivo by an indentation test. Tribology International 39, 12-21.
Pailler-Mattéi, C., Pavan, S., Vargiolu, R., Pirot, F., Falson, F., Zahouani, H., 2007. Contribution of stratum corneum in determining bio-tribological properties of the human skin. Wear 263, 1038-1043.
Pasumarty, S. M., Johnson, S. A., Watson, S. A., Adams, M. J., 2011. Friction of the human finger pad: influence of moisture, occlusion and velocity. Tribology Letters 44, 117-137.
7
Peisker, H., Gorb, S. N., 2010. Always on the bright side of life: anti-adhesive properties of insect ommatidia grating. Journal of Experimental Biolology 213, 3457-3462. Pitenis, A. A., Dowson, D., Sawyer, W. G. 2014. Leonardo da Vinci’s Friction Experiments: An Old Story Acknowledged and Repeated. Tribology Letters 56, 509-515.
Qian, J., Gao, H., 2006. Scaling effects of wet adhesion in biological attachment systems. Acta Biomaterialia 2, 51-58.
Soneda, T., Nakano, K, 2008. Investigation of the vibration detectability of human fingers supported by the observation of contact zones. In: Proceedings of the 2nd international conference on advanced tribology, Singapore iCAT288.
Soneda, T., Nakano, K., 2010. Investigation of vibrotactile sensation of human fingerpads by observation of contact zones. Tribology International 43, 210-217.
Spinner M., Westhoff G., Gorb S. N., 2014. Subdigital setae of chameleon feet: Frictionenhancing microstructures for a wide range of substrate roughness. Scientific Reports 4:5481.
Spurr, A. R., 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. Journal of Ultrastructure Research 26, 31-43.
Tomimoto, M., 2011. The frictional pattern of tactile sensations in anthropomorphic fingertip. Tribology International 44, 1340-1347.
Tomlinson, S. E., 2009. Understanding the friction between human fingers and contacting surfaces. PhD thesis, The University of Sheffield.
Tomlinson, S. E., Lewis, R., Carré, M. J., 2007. Review of the frictional properties of finger– object contact when gripping. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 221, 841-850.
Tomlinson, S. E., Lewis, R., Liu, X., Texier, C., Carré, M. J., 2011. Understanding the friction mechanisms between the human finger and flat contacting surfaces in moist conditions. Tribology Letters 41, 283-294.
8
Van der Heide, E., Zeng, X., Masen, M. A., 2013. Skin tribology: science friction? Friction 1, 130-142.
van Kuilenburg, J., Masen, M. A., van der Heide, E., 2013. A review of fingerpad contact mechanics and friction and how this affects tactile perception. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 0, 1-16.
Warman, P. H., Ennos, A. R., 2009. Fingerprints are unlikely to increase the friction of primate fingerpads. Journal of Experimental Biology 212, 2016-2022.
Wolfram, L. J., 1983. Friction of skin. Journal of the Society of Cosmetic Chemists 34, 465476.
Fig. 1. Adhesive forces of human fingers. The adhesive forces of the tips of index finger (blue circles) and the thumb (red crosses) under different normal load ranging from 0 to 1000 mN measured on epoxy resin replica of a smooth glass surface (a), and surfaces of a grain size of 1.3 µm (b), 10.8 µm (c), and 55.0 µm (d). Scanning electron micrographs of the test surface are presented on the top left 9
side of each graph. The scale bar - 20 µm. The regression lines for measurements of the index finger (black line) and the thumb (dashed line) were calculated by the LOWESS-method.
Fig. 2. Adhesive forces of the human index finger and the thumb (pooled) on resin replicas of different surface roughness under different load conditions: a) 0-300 mN, b) 300-600 mN c) 600-1000 mN. Boxes and horizontal lines represent the upper and lower quartiles and the median values, respectively. Whiskers show the maximum and minimum values. Bars and asterisks indicate significant difference.
10
Fig. 3. The effect of the normal load. Results of pairwise comparisons (pairwise Wilcoxon signed rank sum test) between the ten 100 mN pressure categories for each test surface: a) smooth surface, b) 1.3 µm grain size, c), 10.8 µm grain size, and d) 55.0 µm grain size. Significance levels are indicated by asterisks (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001).
Fig. 4. Relation between the normal load and the apparent contact area. The increase and decrease of normal load (a) and apparent contact area (b) during the experiment. An increase of the normal load leads to an increase of the apparent contact area.
Fig. 5. A fingertip in contact with a glass surface. A glass plate can be hold by the adhesive forces of the tip of a human index finger. Supplementary
Supplementary Fig.1. An example of typical recorded force curve. 1. Test surface is pressing against the finger tip. The sensor detects an increase of normal load force. 2-3. Test surface is removing from the finger tip and the detected force decreases. 3. The detected force is falling under zero level, indicating the presence of the finger tip adhesion to the test surface. 4. Contact rupture. The mean 11
value of noise (N0) was calculated over 400 ms before the measurement. The normal force corresponds to the difference between the maximum force curve and N0, whereas the adhesive force corresponds to the difference between the minimum of the force curve and N0. Supplementary Fig. 2. Fingerprints and surface morphology of the human finger tip. a) and b) Film of stratum corneum cells, sweat, and sebum (fingerprint) released on a glass plate in the stereo microscope (Leica M205 A, Leica Microsystems, Wetzlar, Germany) under the coaxial illumination. c) and d) SEM image (HITACHI S800, Hitachi High-Technologies Corporation, Tokyo, Japan).of the two-step positive mould of the living human finger. Note droplets of the fluid (SG) coming out of the pores of sweat glands. RG, ridges of the finger skin.
12
