Wear 263 (2007) 10161022

Effect of bulk deformation on rubber adhesion

K. Maeda, A. Bismarck, B. Briscoe
Department of Chemical Engineering, Imperial College London, South Kensington Campus, SW7 2AZ, UK
Received 18 August 2006; received in revised form 6 February 2007; accepted 7 February 2007
Available online 23 May 2007

Abstract
In general, rubber friction is divided into two parts; the bulk hysteresis and the contact adhesive term. These two contributions are regarded to be
independent of each other, but this is only a simplified assumption. If the adhesive force is solely a function of the surface free energy, it has been
assumed that this adhesive force per unit area should be constant during any bulk (surface) deformation. The surface free energy is a function of
both internal energy and entropy, and so it should change if the internal energy and/or entropy are changing due to any bulk deformation. In order to
ascertain the effect of the bulk deformation on rubber adhesion, this effect on the surface free energy of two rubber compounds filled with various
loading fractions of carbon black, natural rubber (NR) and styrene butadiene rubber (SBR) was studied. The surface free energy was determined
from measured contact angles on rubber samples at constant applied strain. It was found that the surface free energy increases significantly upon
applying a deformation. However, with increasing the carbon black loading fraction the surface tension increase becomes less pronounced. The
effect of the surface free energy on the surface damage generated by scratch tests was also discussed. It was found that roll debris formation occurs
only in the case of the rubber compounds whose apparent surface free energy increases greatly after applying large bulk deformations.
2007 Elsevier B.V. All rights reserved.
Keywords: Roll formation; Surface energy; Wear mechanism

1. Introduction
Improvement in the durability of rubber compounds is one
of the most important research areas in order to improve the
quality of tyres. The durability of the tyre tread is generally
thought to be affected by the wear resistance of the tread rubber
compound. Since rubber wear coincides with rubber friction, it
is very important to ascertain the corresponding mechanism of
rubber friction. In general, rubber friction is divided into two
parts; the bulk hysteresis and the contact adhesive term [1]. It is
accepted that the rubber friction is a function of the viscoelastic
properties of the bulk rubber [2,3]. Hysteresis friction is considered to be caused by the bulk deformation that occurs during
friction, i.e. the deformational loses [4], which is a function of the
bulk viscoelastic properties of the rubber. The origin of adhesive
friction, however, is much more complicated not least because of
the surface roughness of the counterparts and rubber materials.
Grosch found that the maximum of the adhesive component
of rubber friction on a smooth surface occurs at a sliding veloc-

Corresponding author.
E-mail address: b.briscoe@ic.ac.uk (B. Briscoe).

0043-1648/$ see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.wear.2007.02.002

ity, which correlates to the reduced frequency corresponding to

the maximum of the loss modulus [2]. Ludema and Tabor considered the adhesion of rubber compounds to be a combination
of the true contact area and the shear strength within the thin surface layer [3]. Schallamach explained the dependency of rubber
friction on the viscoelastic properties by applying a thermally
activated rate process model [5]. His calculation indicates that
the adhesion component of rubber friction is a function of the
true contact area as he demonstrated by model experiments [6].
Schallamach observed that a tangential force, which is applied
to a hemispherical rubber sample on a clean smooth surface,
causes the contact surface of the rubber to buckle and generate waves of detachment [7]. It was suggested that energy must
be supplied to the system to keep the waves moving along the
surface to overcome the peeling energy losses.
Briggs and Briscoe showed that the work of adhesion in a
peeling process is the main energy loss [8] and is a function
of peeling velocity and surface roughness [9]. The dependency
of the work of adhesion on the peeling velocity suggests that
the adhesion component of rubber friction may be a function of
the viscoelastic properties. Ideally the adhesion of rubber compounds can be considered as a function of surface free energy,
if the surface damage and bulk damage could be avoided [10].

K. Maeda et al. / Wear 263 (2007) 10161022

In this case, the rubber adhesion is thought to be a reversible

process [11]. Thus, the rubber adhesion is considered to be a
function of the viscoelastic bulk materials properties and the
surface free energy.
These two processes of rubber friction are often thought
to be independent of each other, but this is only a simplified
assumption. The viscoelastic dissipation should be caused by
bulk/interface deformation, and so it seems to suggest that both
adhesion and deformation occur at the same time as response to
the same external stimulus. The peeling or shear stress seems
to be independent of the bulk deformation, which affects the
hysteresis loss because of the different volume of the corresponding region during the deformation. However, in this case
it is based on the assumption that the strength of the junction
between the rubber surface and the counterpart is constant during the deformation. Gent and Petrich suggested that a better
correlation between the experimental peel strength and the theoretical strength calculated from the experimental stressstrain
curve can be obtained if the value of the strength of the interaction between the rubber surface and the counterface could
be varied with the deformation rate instead assuming it to be
constant during the bulk deformation [12].
The surface free energy s at constant pressure and volume
is described as;


 
 
 
G
A
U
S
s =
=
=
T
(1)
s T
s T,V
s T,V
s T,V

1017

Table 1
Composition of materials (parts by weight)
Series of NR

Series of SBR

Natural rubber
Styrene-butadiene (24/76)

100.06

Carbon N234

0, 15, 30, 45, 60

Aromatic oil
Stearic acid
Zinc oxide

10
3
5

37.5
1
3

Sulfer
N-tert-Butyl-2-benzothiazolyl
sulfenamide
Dibenzothiazyldisulfide

2.5
0.5

1.5

100

1.5

the composition of the various rubber samples. The rubber sheets

of thickness 2.0 mm were cut into 2.0 cm 7.0 cm rectangular
specimens. Prior to any measurements the surfaces of the rubber samples were carefully cleaned using diethyl ether (99.9%,
Aldrich) and then dried [14]. In the cases of the prestretched
samples, they were cleaned by wiping using diethyl ether soaked
fluff-free tissue paper after stretching in order to remove additives, which might have migrated from the bulk to the surface.
2.2. Contact angle measurements

2. Experimental

Low-rate dynamic contact angle measurements on sessile

drops were measured using Drop Shape Analyser (DSA 10
MK2, Kruss GmbH, Hamburg, Germany). Only the advancing contact angles were determined. The advancing mode, i.e.
wetting of dry surface, was realized by putting sessile water
droplets of about 7 L on the rubber surfaces and increasing the volume of liquid in the drop by using a motor driven
microsyringe. The rate of droplet expansion was 10 L/min.
At least 10, but more often 20, independent measurements at
different sites of several samples were taken. The largest and
smallest contact angle values in each series were discarded and
standard deviations obtained. The contact angles were measured using deionised water, formamide (99.5+%, Aldrich) and
diiodomethane (99%, Aldrich). All measurements were conducted in an air-conditioned room at 20 C.
Contact angle measurements were also performed on uniaxially prestretched rubber compounds in constant strain mode.
The specimens were stretched by 100, 300 and 700% strain,
respectively. The samples were stretched by a bench vice. Once
the desired strain was adjusted the vice was fixed. The major
difficulty is the polymer relaxation occurring in the elongated
rubber specimens. However, in a rubber compound the polymer main chains are locked in place by covalent crosslinks
formed during the vulcanisation process, which leads to rather
long relaxation times so we assumed that the applied strain is
constant.

2.1. Materials

2.3. Determination of the surface free energy

All rubber compounds used in this research were kindly supplied by Bridgestone Corporation (Japan). Table 1 summarises

Solid surface free energies are commonly estimated using

surface tension models from measured contact angles of series

where G is the Gibbs free energy, A the Helmholtz free energy,

U the internal energy, T the absolute temperature, S the entropy,
V the volume and s the surface area. Eq. (1) shows the surface
free energy is a function of both internal energy and entropy.
The elastic force of rubber compounds is also considered to
be a function of both internal energy and entropy. If a rubber
compound is uniaxially deformed at the same temperature, the
responding force F is;
 
 
 
A
U
S
F=
=
T
(2)
l T
l T
l T
where l is the stretched length. Eq. (2) links both the internal
energy U and entropy S with the material deformation produced
by an applied stress. This implies that the surface free energy s
of a substrate should be a function of the imposed strain. Clarifying the effect of bulk deformation on the surface free energy
of rubber materials might help to explain the rubber friction.
We discuss the effect of bulk deformation on the surface free
energy of rubber compounds and its relation to the surface damage mode. We are also aiming to relate the change in the surface
free energy of rubber compounds to the roll debris formation (as
defined in [13]) during scratching friction.

1018

K. Maeda et al. / Wear 263 (2007) 10161022

of test liquids with known liquid surface tension (components

and parameters) on a solid. The acid/base approach, introduced
by van Oss [15], was used. This approach separates physical
from chemical interactions. The physical interaction component
of the surface energy LW describes the van der Waals interactions (LWLifshitz/van der Waals). The chemical interaction
or acid/base component AB covers Lewis-acid/base (including
proton donor/acceptor) interactions as well as hydrogen bonding.
Critics of the surface tension component approaches stress
that there is no thermodynamic basis to the separation approach
[16,17]. However, distinguishing the chemical interaction from
the physical one was not important for this work.
2.4. Scratch tests
The pendulum scratching experiments were reported elsewhere [1820]. Scratch tests quickly provide various samples
with adequate surface damage. In order to ascertain the effect of
the bulk deformation on the surface damage mode, the pendulum
scratching test was performed using a sharp needle whose semi
cone angle was 15 because the strain of the bulk deformation
is thought to be larger if the tip of the needle is sharper. The
scratching tests were repeated 10 times with the initial angle of
150 for each rubber compound. The experimental details are
fully described in [20].
3. Results and discussion
3.1. Contact angle and surface free energy
In order to determine the effect of bulk deformation on rubber adhesion, the surface free energies of both unstrained and
uniaxially strained rubber samples were determined from measured contact angles using the acid and base approach [15]. As
was suggested in the introduction, both the internal energy and
entropy of a rubber compound are affected by the bulk deformation, and so it was assumed that the surface free energy has to
change too. The apparent surface free energy of the natural rubber (NR) compounds is larger than that of the styrene-butadiene
rubber (SBR) compounds, and the apparent surface free energy
was increased if the imposed elongation strain was increased.
The apparent surface free energy is plotted against the volume
of the carbon black (CB) loaded into the rubber compounds. The
95% probability based upon a standard normal distribution with
a 1.96 standard deviation is taken as error. Fig. 1 displays the
results for a series of the NR compounds. The apparent surface
free energy is very similar for all the non-stretched NR compounds (21.1 1.3 mJ/m2 ). The apparent surface free energy of
the rubber increases significantly with increasing applied strain.
However, the effect of the applied strain on the surface free
energy decreases with increasing the CB loading fraction. The
presence of the CB in the NR affects the deformation of the
polymer molecules because of its reinforcement effects, i.e. the
interaction between the CB and the NR. It seems that the CB
obstructs the extension of the polymer main chains in the loaded
volume, which results in a smaller increase of the apparent sur-

Fig. 1. The surface free energy of the natural rubber (NR) compounds with the
volume of loaded black carbon for static and elongated samples. Symbols: ()
non-stretched; () 100% stretched; () 300% stretched; () 700% stretched.
Error bar means 95% probability based on the standard normal distribution.

face free energy. In fact, the apparent surface free energy in

the case of 60 phr CB shows little difference from the original
(ca. 0.6 mJ/m2 ) even though the sample was stretched to 700%
strain. The observation that the determined surface free energy
increases with increasing imposed uniaxial strain seems to support the assumption that the surface free energy is affected by
the changing the internal energy and the entropy of the bulk
material.
Fig. 2 illustrates the results for a range of the SBR compounds.
In the case of the SBR samples strained by 100% no significant
difference in the apparent surface energy can be found. The SBR
compound with 60 phr CB has a significant difference from the
original, but the error bars overlap. Detectable differences in
the apparent surface free energy are found in the case of the
pure SBR compounds the 15 phr CB loaded SBR if they are
elongated to 300% strain. It may be due to the same reason as
the case of the NR compounds; the extension of the polymer
chains is obstructed by the reinforcement of the loaded CB.

Fig. 2. The surface free energy of the styrene-butadiene rubber (SBR) compounds with the volume of loaded black carbon for static and elongated samples.
Symbols: () non-stretched; () 100% stretched; () 300% stretched; ()
700% stretched. Error bar means 95% probability based on the standard normal
distribution.

K. Maeda et al. / Wear 263 (2007) 10161022

1019

3.2. Implications for roll debris formation

Stretching a rubber affects its surface free energy and, therefore, the adhesion between the rubber and the counterface should
change. As a consequence, the bulk deformation alters the contribution of the rubber adhesion friction to the total friction, which

Fig. 3. Contact angle hysteresis of SBR compound loaded 60 phr CB with formamide as a function of strain of uni-axial stretching. Symbols: () advancing
angle; () receding angle.

Fig. 3 shows the advancing and receding contact angles measured using droplets of the partially wetting liquid formamide
on the SBR compound containing 60 phr CB. As can be seen
the contact angles clearly decrease with increasing extension of
the rubber compound. It indicates an improved wettability of
the rubber surfaces, which is most likely due to the increase of
the surface free energy of the rubber compound caused by the
decrease in the entropy of the rubber. The contact angle hysteresis, i.e. the difference between the advancing and receding
contact angles, however, is not affected by the deformation of the
sample. It is commonly assumed that the contact angle hysteresis
is due to chemical surface inhomogeneities or surface roughness or time-depended effects such as swelling and absorption
[21,22]. The experimental fact that the contact angle hysteresis remains constant ( = 40 ) indicates that both the extent of
the surface roughness as well as the chemical heterogeneities of
the sample remain unaffected by the applied deformation. The
change of the contact angle which is produced by the bulk deformation points to the assumption that the bulk deformation may
cause changes in the adhesion component of rubber friction and
in the mechanism of rubber wear.

Fig. 4. Frictional forces of NR and SBR compounds during pendulum scratching

with the needle of 15-degree half cone angle as function of the volume of loaded
CB. Symbols: () NR compounds; () SBR compounds.

Fig. 5. Optical microscopic images for NR compounds after scratching by the

needle with 15.0 half cone angle (magnification 10). (a) Non-loaded CB; (b)
15 phr CB; (c) 30 phr CB.

1020

K. Maeda et al. / Wear 263 (2007) 10161022

in turn should affect the surface damage mode. The surface free
energy is one of the key factors in defining adhesive interactions
[10], and so it is likely that rubber wear due to the rubber adhesion may be enhanced by the larger bulk deformation in this
case.
The NR compounds and the SBR compounds whose quasistatic mechanical properties are very similar were selected. Their
frictional behaviour and the surface damage produced by pendulum scratching were investigated [20]. The pure NR compound
shows the roll debris formation after scratching whereas the SBR
with 45 phr CB does not show any roll formation, although the
stressstrain curves of two compounds are very similar. The
reason for this difference could not be explained because of
the complexity of the mechanism of rubber wear. However, it
can be explained by the fact that the bulk deformation during
the scratching event leads to a drastic increase of the apparent
surface free energy, which will in turn affect the adhesive interaction between the rubber surface and the counterface. The results
of pendulum scratching of NR and SBR compounds using 15
semi cone angle of the stylus are shown in Fig. 4. For both rubber
compounds the friction force increases with increasing the CB
loading fraction. There is hardly any difference between the two

rubber compounds in terms of their frictional behaviour. Fig. 5

shows optical micrographs of the surface damage after a single scratch test for the NR compounds. The classical roll debris
formation was found to occur during scratching of unfilled NR
compound (Fig. 5a). The extent of the roll debris formation of
the NR compounds decreases with increasing the CB loading
fraction (Fig. 5b) and it disappeared almost completely for the
NR containing 30 phr CB (Fig. 5c). The roll debris formation
is clearly a function of the CB volume loading fraction. The
tensile strength of the rubbers increases with increasing loading
fraction of the CB. The increased strength may affect the roll
debris formation, because it is harder to peel of a thin layer of
the elastomer surface.
The SBR compounds, whose mechanical strength is very
similar to that of the NR compounds [20], should show the roll
formation after a single scratching if it would be true that the roll
debris formation is solely determined by the mechanical strength
of the rubber bulk/surface. Fig. 6 shows the optical micrographs of the damaged SBR surfaces. However, even though the
investigated SBR samples (45 and 60 phr CB) have very similar quasi-static mechanical properties as compared to the NR
compounds (unfilled and 15 phr CB, respectively), the observed

Fig. 6. Optical microscopic images for SBR compounds after scratching by the needle with 15.0 cone angle (magnification 10). (a) Non-loaded CB; (b) 30 phr
CB; (c) 45 phr CB; (d) 60 phr CB.

K. Maeda et al. / Wear 263 (2007) 10161022

surface damage (Fig. 6c and d) is not comparable to that of

the NR compounds. For the SBR compounds with even lower
mechanical strength, i.e. containing less than 45 phr CB, no roll
debris formation was observed (Fig. 6a and b). This observation
suggests that the strength of the materials may not be the only
factor affecting the roll debris formation.
The observed increase in the surface free energy of the rubber
compounds with increasing uniaxial strain provides some useful insights. The roll debris formation was more prominent for
the NR compounds for which the apparent surface free energy
increased more drastically when stretched to 700% in comparison with the SBR compounds. This finding suggests that a
correlation between the change of the surface free energy at
high strain and the adhesive friction term, which affects the roll
debris formation, exists. The contribution of the surface free
energy to the total rubber friction, however, is generally rather
small, as can be seen by the fact that the surface free energies
of the NR and SBR samples at low applied strains are rather
similar. For example, the surface free energy of the NR containing 15 phr CB is 23.0 1.1 mJ/m2 and that of the SBR with
45 phr CB is 22.2 1.5 mJ/m2 at an applied strain of 100%. Furthermore, the quasi-static mechanical properties are also very
similar [20]. However, at an applied strain of 700% the apparent
surface free energy of the NR containing 15 phr CB increases
to 31.8 3.4 mJ/m2 whereas of the SBR containing 60 phr CB
remains almost constant at 25.7 1.7 mJ/m2 .
Our result suggests that the roll debris formation results
from the strong adhesive interaction between the elastomer
surface and the counterface, which will be affected by the
large surface/bulk deformation arising during scratching. This
suggestion is supported by the observed increase in the surface free energy of the rubbers upon experiencing an applied
strain.
4. Conclusion
The effect of an imposed uniaxial strain on the surface free
energy of pure and the carbon black (CB) loaded natural rubber and styrene-butadiene rubber compounds was investigated in
order to determine the effect of the bulk deformation on adhesive
friction. The surface free energy was estimated from measured
contact angles of test liquids on prestretched rubber surfaces.
It was found that the computed surface free energy of the CB
filled NR and SBR compound increased with increasing applied
uniaxial strain. The measured contact angle hysteresis remained
constant with increasing strain, and so it can be assumed that
a change of the surface roughness or surface heterogeneity can
be excluded. It could be inferred that the increase of the apparent surface free energy may be due to a change of both the
internal energy and entropy as a consequence of the bulk deformation.
These results suggest that the adhesive term of rubber friction is affected by the surface free energy of rubber compounds,
which increases with increasing the bulk deformation. It is contrary to the common assumption, but it could be speculated that
the rubber adhesion may be connected to the hysteresis friction. In this case the surface damage mode will be affected by

1021

the bulk deformation, because the contribution of the adhesion

to the total rubber friction increases at a high strain as highlighted by the increased surface free energy. We found that the
roll debris formation seems to correlate with the relative increase
of the computed surface free energy as a function of the imposed
uniaxial strain. The roll debris formation may require the strong
interaction between the elastomer surface and the counterface
in order to localise the deformation of the thin layer at the
contact.
This idea to investigate the connection between the bulk
deformation and the surface adhesive friction may help to provide a better understanding of rubber friction, which may lead
to improved rubber compounds with better durability.
Acknowledgement
K.M. is very grateful to Bridgestone Corporation for providing his scholarship.
References
[1] D. Tabor, Hysteresis losses in the friction of lubricated rubber, Rubber
Chem. Technol. 33 (1960) 142150.
[2] K. Grosch, The relation between the friction and viscoelastic properties of
rubber, Rubber Chem. Technol. 37 (1964) 386403.
[3] K. Ludema, D. Tabor, The friction and visco-elastic properties of polymeric
solids, Wear 9 (1966) 329348.
[4] J. Greenwood, D. Tabor, The friction of hard sliders on lubricated rubber,
Proc. Phys. Soc. 71 (1958) 9891001.
[5] A. Schallamach, A theory of dynamic rubber friction, Wear 6 (1963)
375382.
[6] A. Schallamach, Friction and Wear, Wear 1 (1957/58) 384417.
[7] A. Schallamach, How does rubber slide? Wear 17 (1971) 301312.
[8] G. Briggs, B. Briscoe, The dissipation of energy in the friction of rubber,
Wear 35 (1975) 357364.
[9] G. Briggs, B. Briscoe, Surface roughness and the friction and adhesion of
elastomers, Wear 57 (1979) 269280.
[10] K. Johnson, K. Kendall, A. Roberts, Surface energy and the contact of
elastic solids, Proc. R. Soc. London A 324 (1971) 301313.
[11] S. Wu, Polymer Interface and Adhesion, Marcel Dekker Inc., New York,
1982 (pp. 339342).
[12] A.N. Gent, R.P. Petrich, Adhesion of viscoelastic materials to rigid substrates, Proc. R. Soc. London A310 (1969) 433448.
[13] S. Aharoni, Wear of polymers by roll-formation, Wear 25 (1973) 309
327.
[14] C. Extrand, A. Gent, Contact angle and spectroscopic studies of chlorinated and unchlorinated natural rubber surfaces, Rubber Chem. Technol.
61 (1988) 688697.
[15] C.J. van Oss, Use of the combined Lifshitz-van der Waals and Lewis acidbase approaches in determining the apolar and polar contributions to surface
and interfacial tensions and free energies, J. Adhes. Sci. Technol. 16 (2002)
669677.
[16] D. Kwok, D. Li, A. Neumann, Evaluation of the Lifshitz-van der
Waals/acid-base approach to determine interfacial tensions, Langmuir 10
(1994) 13231328.
[17] D. Kwok, Y. Lee, A. Neumann, Evaluation of the Lifshitz-van der
Waals/acid-base approach to determine interfacial tensions 2, Langmuir
14 (1998) 25482553.
[18] B. Lamy, Effect of brittleness index and sliding speed on the morphology
of surface scratching in abrasive or erosive processes, Tribol. Int. 17 (1984)
3538.
[19] B. Briscoe, A. Delfino, E. Pelillo, Single-pass pendulum scratching
of poly(styrene) and poly(methylmethacrylate), Wear 225229 (1999)
319328.

1022

K. Maeda et al. / Wear 263 (2007) 10161022

[20] K. Maeda, A. Bismarck, B. Briscoe, Mechanisms of scratching frictions

and damage maps for rubber compounds, Wear 259 (2005) 651660.
[21] R.E. Johnson Jr., R.H. Dettre, Contact Angle Hysteresis. I. Study of an
Idealized Rough Surface, in: Contact Angle Wettability and Adhesion,
Advances in Chemistry Series 43, American Chemical Society, Washington, D.C., 1964 (pp. 112135).

[22] R.H. Dettre, R.E. Johnson Jr., Contact Angle Hysteresis. II. Contact Angle
Measurements on Rough Surfaces, in: Contact Angle Wettability and
Adhesion, Advances in Chemistry Series 43, American Chemical Society,
Washington, D.C., 1964 (pp. 136144).