We have measured the surface topography and calculated the surface roughness power spectrum for an asphalt road surface. For the same surface we have measured the friction for a tire tread compound for velocities 10(-6) m s(-1) < v < 10(-3) m s(-1) at three different temperatures (at -8 °C, 20 °C and 48 °C). The friction data was shifted using the bulk viscoelasticity shift factor a(T) to form a master curve. We have measured the effective rubber viscoelastic modulus at large strain and calculated the rubber friction coefficient (and contact area) during stationary sliding and compared it to the measured friction coefficient. We find that for the low velocities and for the relatively smooth road surface we consider, the contribution to friction from the area of real contact is very important, and we interpret this contribution as being due to shearing of a very thin confined rubber smear film.

The energy dissipation in the contact regions between solids in sliding contact can result in high local temperatures which may strongly effect friction and wear. This is the case for rubber sliding on road surfaces at speeds above 1 mm s(-1). We derive equations which describe the frictional heating for solids with arbitrary thermal properties. The theory is applied to rubber friction on road surfaces and we take into account that the frictional energy is partly produced inside the rubber due to the internal friction of rubber and in a thin (nanometer) interfacial layer at the rubber-road contact region. The heat transfer between the rubber and the road surface is described by a heat transfer coefficient which depends on the sliding speed. Numerical results are presented and compared to experimental data. We find that frictional heating results in a kinetic friction force which depends on the orientation of the sliding block, thus violating one of the two basic Leonardo da Vinci 'laws' of friction.

In rubber friction studies it is usually assumed that the friction force does not depend on the sliding direction, unless the substrate has anisotropic properties, like a steel surface grinded in one direction. Here we will present experimental results for rubber friction, where we observe a strong asymmetry between forward and backward sliding, where forward and backward refer to the run-in direction of the rubber block. The observed effect could be very important in tire applications, where directional properties of the rubber friction could be induced during braking.

In rubber friction studies it is often observed that the kinetic friction coefficient µ depends on the nominal contact pressure p. We discuss several possible origins of the pressure dependency of µ: (a) saturation of the contact area (and friction force) due to high nominal squeezing pressure, (b) non-linear viscoelasticity, (c) non-randomness in the surface topography, in particular the influence of the skewness of the surface roughness profile, (d) adhesion, and (e) frictional heating. We show that in most cases the non-linearity in the µ(p) relation is mainly due to process (e) (frictional heating), which softens the rubber, increases the area of contact, and (in most cases) reduces the viscoelastic contribution to the friction. In fact, since the temperature distribution in the rubber at time t depends on on the sliding history (i.e., on the earlier time t ′ < t), the friction coefficient at time t will also depend on the sliding history, i.e. it is, strictly speaking, a time integral operator.The energy dissipation in the contact regions between solids in sliding contact can result in high local temperatures which may strongly affect the area of real contact and the friction force (and the wear-rate). This is the case for rubber sliding on road surfaces at speeds above 1 mm/s. In Ref. [14] we have derived equations which describe the frictional heating for solids with arbitrary thermal properties. In this paper the theory is applied to rubber friction on road surfaces. Numerical results are presented and compared to experimental data. We observe good agreement between the calculated and measured temperature increase.

No abstract