Biotribology 26 (2021) 100179

Contents lists available at ScienceDirect

Biotribology
journal homepage: www.elsevier.com/locate/biotri

How does ‘Gecko tape’ work?

Hans Terwisscha-Dekker *, Marion Grzelka, Simon Lépinay, Daniel Bonn
Van der Waals-Zeeman Institute, Institute of Physics University of Amsterdam, 1098XH Amsterdam, the Netherlands

A R T I C L E I N F O A B S T R A C T

Keywords: Human-made adhesives lose their tack rapidly after first use, while animals such as geckos can reuse their ad­
Adhesion hesive feet for a lifetime. Nature’s use of fibrillar structures as strong, renewable and self-cleaning adhesives has
Gecko tape inspired the development of synthetic adhesives with similarly structured surfaces. More than a decade of
Dry adhesive
research and engineering has culminated in ‘gecko tape’: a re-useable adhesive that has a structured surface
similar to that of geckos and that outperforms the usual sticky tape. We report experiments that show that,
despite its name, a commercial gecko tape shares few adhesive principles with its eponym. In particular, we find
no evidence that the micrometric features that are present on the surface of the gecko tape play a role in its
adhesive strength. In addition, we find that contrary to the gecko, the tape leaves behind a layer of adhesive after
removal from the surface. The fact that the gecko tape outperforms a conventional adhesive tape is due to the fact
that the softness of the backing of the gecko tape allows to create a much larger contact area for a given normal
force. The conclusion is that surface features are not necessary to create a superb adhesive; tuning the backing
layer elasticity may be enough.

1. Introduction stress distribution, and a soft interface, providing a large contact area,
seems to be a universal design motif, whether the adhesive surface is
Animals can climb many surfaces due to the high friction that their structured or not [15,16]. In fact, the gecko also uses this strategy, as its
feet experience. A large part of the friction is adhesive; the exceptional largest surface structures are most stiff, and the smallest features most
adhesive abilities found in the animal kingdom [1], have inspired ma­ compliant [16,17].
terial scientists for years to develop biomimetic adhesives [2,3]. The The prospect of being able to create an all-purpose reusable adhesive
poster boy for these so-called ‘dry adhesives’ is the gecko, an animal that has led to the commercialisation of the gecko adhesion mechanisms into
can climb onto almost any surface due to the sophisticated adhesive a product marketed as ‘gecko tape’ (Fig. 1A). Among different manu­
fibrillar structures found on its feet[4]. The soft fibrils that contact a facturers, it is sold as a few millimetres thick double-sided tape, that is
countersurface ensure that the gecko can make a large contact area with reusable and washable. The question arises, however, whether this tape
almost any surface whether rough or smooth. In this way, geckos can really employs the same adhesive principles as its eponym, the gecko. In
stick due to the weakest and most generic of intermolecular interactions, this paper, we investigate the physics of commercial ‘gecko tape’
the Van der Waals force [4]. However, the exact mechanisms of gecko adhesion by using mechanical testing, microscopy and Raman
adhesion are complex [5,6]: gecko feet have fibrillar features on a hi­ spectroscopy.
erarchy of length scales, and the attachment and detachment processes
depend sensitively on the angle the gecko makes with the substrate [7]. 2. Experiments
To reduce this complexity, researchers started making artificial
structured adhesives from soft rubbers, with a surface covered with The first question that arises is what the gecko tape surface looks like
pillars [2,8,9]. This has allowed systematic study of structural parame­ at micrometric scales. An optical profilometer (Keyence-VX100, 404 nm
ters such as pillar shape [10] and orientation [11], pillar ordering [12] laser, 5× objective) is used to characterise the surface profile. The gecko
and backing layer stiffness [13]. This last factor has emerged to be a tape surface profile shows a dilute (~102 mm− 2), disordered array of
defining factor for adhesive strength, even for smooth, unstructured pillars (Fig. 2A). This is intriguing, since research has shown that both
surfaces [14]. A combination of a stiff backing layer, providing an even disorder [12] and a low coverage of pillars [19] annul the effect of

* Corresponding author.
E-mail address: [email protected] (H. Terwisscha-Dekker).

https://doi.org/10.1016/j.biotri.2021.100179
Received 30 November 2020; Received in revised form 10 February 2021; Accepted 7 March 2021
Available online 17 March 2021
2352-5738/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
H. Terwisscha-Dekker et al. Biotribology 26 (2021) 100179

Fig. 1. Pictures of gecko tape. A A roll of gecko tape from Stickie [18] (l = 3 m, w = 2.9 cm, h = 2 mm). B To remove the fibrillar surface structures, the gecko tape is
cut parallel to the adhesive surface.

Fig. 2. A Optical profilometry image of a millimetric piece of tape shows clearly the distribution of pillars, B which are tens of micrometers high and wide. C Height
profile of cut gecko surface, which is unstructured. D Height profile of conventional tape.

surface structure, leading to an adhesion that is the same as a flat sur­ measure the normal force F as a function of the vertical displacement d.
face. The pillars are tens of micrometers high and wide (Fig. 2B). To disentangle the effects of surface structure and bulk material
To simultaneously measure adhesion forces and image the contact properties, adhesion tests are performed with three surfaces: the original
area, we use a home-built mechanical set up on top of an inverted mi­ gecko tape surface featuring the pillars, the surface of a piece of gecko
croscope (Zeiss Axiovert 200 M, 2.5× objective), shown in Fig. 3. Mil­ tape from which the pillars have been removed (Fig. 1B), and conven­
limetric circles of tape are glued to a screw connected a force sensor tional adhesive tape (Tesa Double-Sided Tape Universal, w = 5.0 cm, h
(Zemic, 5KG) that is mounted on a vertical translation stage. The = 90 μ m). To remove the pillars, ~25% of the original thickness of the
countersurface is a glass microscopy slide (VWR, h = 1 mm) thoroughly gecko tape is cut away with a scalpel. The profile of the cut gecko tape
cleaned with Hellmanex and ethanol. This set up allows us to accurately shows an unstructured surface (Fig. 2C), similar to the profile of the

2
H. Terwisscha-Dekker et al. Biotribology 26 (2021) 100179

conventional tape (Fig. 2D). Adhesion experiments are performed by

separating a pre-loaded tape-glass contact (vload = 20 μ m ⋅ s− 1) with a
constant retraction velocity vretract = 10 μm ⋅ s− 1.
Fig. 4A shows typical force-distance curve for each of the three
studied tapes. We define d = 0 as the point where the probe and the
countersurface start to touch. We quantify the stiffness k by linearly
fitting F = kd to the compressive part of the force-distance curve
(Fig. S2). We find kgecko = 9.26 ± 0.05 ⋅ 102 N ⋅ m− 1 and kconventional =
1.48 ± 0.03 ⋅ 104 N ⋅ m− 1. The cut gecko tape does not have a unique
stiffness (the force-distance relation is non-linear, Fig. S2), but its stiff­
ness is of the same order of magnitude as that of the original gecko tape.
When the maximum compressive force is reached, we image the
tape-glass interface (Fig. 4B). Qualitatively, we can see that the gecko
tape forms a continuous contact with the glass countersurface. This
means that it does not touch the glass at discrete points with its pillars. It
appears that because of the softness of the tape material, the pillars are
strongly deformed upon contacting the countersurface, and continuous
contact develops. Images made at higher resolution (Fig. S3) confirm
these observations.
Quantitatively, the gecko tape generates a significantly larger con­
tact area at a certain compressive force than the conventional tape
(Fig. 4C, details about contact area measurement are in Supplementary
Information). This difference can be attributed to the order-of-
Fig. 3. The tape (orange) is connected to a force sensor which measures the magnitude lower stiffness of the gecko tape compared to the conven­
force F under an imposed vertical displacement d. The tape-glass interface is tional tape, which allows the gecko tape to deform much more upon
imaged using a microscope. contacting the glass slide.
The adhesion mechanics of gecko tape and the conventional tape are

Fig. 4. A Typical force-distance curves. Positive d indicates compression of the surfaces, positive F a compressive force. Vice versa, negative d indicates separation of
the surfaces, where F is tensile. B Microscopy images of tape-glass interface immediately before start of retraction. Black areas indicate tape-glass contact. Striped
patterns are due to interference with glass and probe. C Contact mechanics: contact area A versus normal force F for original gecko tape ●, cut gecko surface ▴ and
conventional tape ■ D Work of adhesion W versus contact area A. The dotted line and the grey shadow indicate the linear fit with the error interval, used to extract
the surface energy W/A.

3
H. Terwisscha-Dekker et al. Biotribology 26 (2021) 100179

Fig. 5. A Residual layer on glass after contact with the gecko tape. B Raman spectrum of different gecko tape surfaces and residual layers. Polyurethane and glass
data are shown for reference. Characteristic polyurethane peaks at 1732, 1445 and 1300 cm− 1 are indicated with dotted lines.

very different (Fig. 4A). The conventional tape is around ten times 3. Discussion and Conclusion
thinner than the gecko tape (hconventional = 90 μ m, hgecko = 1 mm) and
cannot sustain an adhesive force over such a long distance as the gecko The surface of commercial gecko tape is covered with micrometric
tape. We quantify adhesive strength with the work of adhesion: W = − pillars, suggesting its structured surface contributes to generating strong
∫ d*
adhesion, just like for the gecko. However, the observed density of the
− ∞ Fretract δd. d* is the distance at which the force starts to be negative
(tensile) during retraction. pillars is very low, 102 times lower than gecko setae density [4]. Adhe­
When looking at the work of adhesion W as a function of contact area sion experiments show that the presence of the pillars does not increase
A (Fig. 4C), it is remarkable that all the measurements of work of gecko tape adhesion; adhesion strength of the original gecko tape is very
adhesion W as a function of the contact area A follow a straight line that similar to that of gecko tape from which the pillars have been removed.
goes through the origin (dotted line in Fig. 4D). This means the effective In addition, in microscopy images of gecko tape-glass interfaces, no
adhesive surface energy is constant, and can be estimated from the fit to discrete contact of only the pillars was observed. In all cases, a contin­
be W/A = 4.0 ± 0.5 ⋅ 102 N ⋅ m− 1. uous contact area can be seen, which implies the pillars play little role in
Apparently, the pillars on the original gecko tape do not change the the contact mechanics.
surface energy as compared to an unstructured surface or one covered Nevertheless, a piece of gecko tape features a much higher work of
with adhesive. adhesion than a piece of conventional adhesive tape of similar size. This
It is also important to note that, in the adhesion experiments, the is because it can generate a much larger contact area under the same
gecko tape does not really qualify as a ‘dry adhesive’: a residual layer of normal force. In this way, it does resemble the gecko, which generates a
material of around 10 μm thick remains at the glass countersurface after large contact area with its soft fibrils.
contact (Fig. 5A). To understand what this layer is made of, we perform The work of adhesion per unit area, an effective ‘surface energy’, of
Raman spectroscopy (WITec UHTS 300, 532 nm excitation laser) on the the gecko tape is remarkably similar to that of the conventional tape,
original tape surface, the cut surface and the residual layers on glass as which has a surface covered with glue. Furthermore, the gecko tape
resulting from contact with the original surface or the cut surface leaves a residual layer after contacting a countersurface, so we can
(Fig. 5A). The peaks at 1732, 1445 and 1300 cm− 1 are characteristic for conclude that the gecko tape also uses glue to stick instead of mere Van
polyurethane [20]: they are indeed also present in the spectrum of a der Waals interactions between dry surfaces as the gecko does. Raman
polyurethane reference (Selectophore™, Merck). Thus, we can confirm measurements leave the possibility open that the glue is made up of
the gecko tape is made of polyurethane (as indicated by the manufac­ unreticulated polyurethane polymers.
turer). The 1732 cm− 1 peak originates from a carbonyl group, which can To conclude, gecko tape is not very gecko-like; its pillars do not play
be either part of the polyurethane amide group, or the isocyanate group any significant role in its adhesion strength and it uses a kind of glue to
of a precursor. Since this peak is present in the gecko tape and its res­ bond to a countersurface. Only the use of a soft surface to generate a
idue, but not in the polyurethane reference, this suggests that the gecko large contact area is a principle that it shares with the gecko. A
tape contains unreacted precursor molecules. When this precursor is a recommendation we can extend to gecko tape manufacturers is to spare
polymer, it probably acts as glue. the effort of creating micrometric structures, and focus on finding the
optimal stiffness and chemistry for creating durable, reusable and strong
adhesive contact.

4
H. Terwisscha-Dekker et al. Biotribology 26 (2021) 100179

Author Contributions [6] Michael Varenberg, et al., Geometry-controlled adhesion: Revisiting the contact
splitting hypothesis, in: Applied Physics A: Materials Science and Processing 103.4,
2011, pp. 933–938, ISSN: 09478396, https://doi.org/10.1007/s00339-011-6394-0.
H.T.-D. conducted the profilometer, adhesion and microscopy ex­ [7] Kellar Autumn, Nick Gravish, Gecko adhesion: evolutionary nanotechnology, in:
periments. H.T.-D. and S.L. conducted the Raman experiments. H.T.-D., Philosophical Transactions of the Royal Society A: Mathematical, Physical and
M.G., S.L. and D.B. interpreted the data and wrote the manuscript. Engineering Sciences, 2008, pp. 1575–1590, 366.1870. issn: 1364503X, https://d
oi.org/10.1098/rsta.2007.2173.
[8] Liehui Ge, et al., Carbon nanotube-based synthetic gecko tapes, in: Proceedings of
Funding the National Academy of Sciences 104.26, Publisher: National Academy of Sciences
Section: Physical Sciences, 2007, pp. 10792–10795, issn: 0027–8424, 1091–6490,
https://doi.org/10.1073/pnas.0703505104.
This project has received funding from the European Research [9] Ming Zhou, et al., Recent advances in gecko adhesion and friction mechanisms and
Council (ERC) under the European Union’s Horizon 2020 research and development of gecko-inspired dry adhesive surfaces, in: Friction 1.2, 2013,
innovation programme (Grant agreement No. 833240). pp. 114–129, ISSN: 22237704, https://doi.org/10.1007/s40544-013-0011-5.
[10] Craig K. Hossfeld, et al., Detachment behavior of mushroom-shaped fibrillar
adhesive surfaces in peel testing, in: Langmuir 29.49, 2013, pp. 15394–15404, ISSN:
Declaration of Competing Interest 07437463, https://doi.org/10.1021/la402838y.
[11] Burak Aksak, Michael P. Murphy, Metin Sitti, Adhesion of biologically inspired
vertical and angled polymer microfiber arrays, in: Langmuir 23.6, 2007,
The authors declare that they have no known competing financial pp. 3322–3332, ISSN: 07437463, https://doi.org/10.1021/la062697t.
interests or personal relationships that could have appeared to influence [12] Henriëtte E. Bakker, Stefan B. Lindström, Joris Sprakel, Geometryand rate-
the work reported in this paper. dependent adhesive failure of micropatterned surfaces, in: Journal of Physics
Condensed Matter 24.6, 2012, https://doi.org/10.1088/0953-8984/24/6/065103
ISSN: 09538984.
Appendix A. Supplementary data [13] Sukho Song, et al., Controllable load sharing for soft adhesive interfaces on three-
dimensional surfaces, in: Proceedings of the National Academy of Sciences 114.22,
Supplementary data to this article can be found online at https://doi. 2017, https://doi.org/10.1073/pnas.1620344114. E4344–E4353. issn:
0027–8424.
org/10.1016/j.biotri.2021.100179. [14] Michael D. Bartlett, et al., Looking beyond fibrillar features to scale geckolike
adhesion, in: Advanced Materials 24.8, 2012, pp. 1078–1083, ISSN: 09359648,
References https://doi.org/10.1002/adma.201104191.
[15] Daniel R. King, et al., Creating gecko-like adhesives for “real world” surfaces, in:
Advanced Materials 26.25, 2014, pp. 4345–4351, ISSN: 15214095, https://doi.
[1] E. Arzt, S. Gorb, R. Spolenak, From micro to nano contacts in biological attachment org/10.1002/adma.201306259.
devices, in: Proceedings of the National Academy of Sciences 100.19, 2003, [16] A.G. Gillies, et al., Gecko toe and lamellar shear adhesion on macroscopic,
pp. 10603–10606, ISSN: 0027-8424, https://doi.org/10.1073/pnas.1534701100. engineered rough surfaces, in: Journal of Experimental Biology 217.2, 2014,
[2] D. Brodoceanu, et al., Hierarchical bioinspired adhesive surfaces-A review, in: pp. 283–289, issn: 0022–0949, https://doi.org/10.1242/jeb.092015.
Bioinspiration and Biomimetics 11.5, 2016, https://doi.org/10.1088/1748-3190/ [17] TaeWan Kim, Bharat Bhushan, Adhesion analysis of multi-level hierarchical
11/5/051001 ISSN: 17483190. attachment system contacting with a rough surface, in: Journal of Adhesion Science
[3] Jeffrey Eisenhaure, Seok Kim, A review of the state of dry adhesives: biomimetic and Technology 21.1, 2007, pp. 1–20, ISSN: 01694243, https://doi.org/10.1163
structures and the alternative designs they inspire, in: Micro-Machines, 2017, /156856107779976097.
pp. 1–38, 8.4. issn: 2072666X, https://doi.org/10.3390/mi8040125. [18] Stickie tape, URL: https://stickietape.nl/.
[4] K. Autumn, Mechanisms of adhesion in geckos, in: Integrative and Comparative [19] Laëtitia Dies-Diverchy, Influence d’une texturation déformable sur l’adhésion et la
Biology 42.6, 2006, pp. 1081–1090, issn: 1540-7063, https://doi.org/10.1093/icb/ friction. PhD Thesis, Université Paris Sud, 2015.
42.6.1081. [20] Parnell Shane, K. Min, M. Cakmak, Kinetic studies of polyurethane polymerization
[5] Marleen Kamperman, et al., Functional adhesive surfaces with “Gecko” effect: the with Raman spectroscopy, in: Polymer 44.18, 2003, pp. 5137–5144, ISSN:
concept of contact splitting, in: Advanced Engineering Materials 12.5, 2010, 00323861, https://doi.org/10.1016/S0032-3861(03)00468-3.
pp. 335–348, ISSN: 14381656, https://doi.org/10.1002/adem.201000104.