Comparison of Multiscale Analytical Model of Friction and Wear of Viscoelastic Materials With Experiments

Friction and wear of viscoelastic materials like rubbers are topics of extreme practical importance such as the construction of tires, shoe heels and soles, rubber O-ring seals, and wiper blades. Friction of viscoelastic materials differs from the frictional properties of the elastic solids as friction is directly related to energy dissipation via the internal damping of such materials while purely elastic materials do not dissipate energy. Based on hysteresis properties of viscoelastic materials, physics based multiscale models were developed by Persson for fiction [1, 2] and powdery wear [3] of rubbers sliding on rough surfaces. In this research, these theories were studied and the theoretical results were compared with experimental results obtained from a dynamic friction/wear tester. The inputs to the theoretical models were the fractal properties of the rough surface, the dynamic modulus, and the fatigue behavior of the viscoelastic material. The fractal properties of the rough surface was obtained from the 3D profile of the surface measured using an optical profilometer. The dynamic modulus of the rubber samples was characterized via dynamic mechanical analysis at different frequencies and temperatures. The fatigue crack growth behavior of the samples were found from experimental results of crack propagation versus tearing energy obtained from the fatigue test. Then, the friction coefficient between different rubber samples and rough surfaces was calculated as a function of sliding velocity using both analytical model and experimental approach. In the dynamic friction/wear tester, normal force was adjusted and measured accurately, in addition, the frictional force was measured using a load cell in longitudinal direction along the sliding axis. The experimental sliding friction coefficient was calculated as the ratio of longitudinal force at a constant velocity to the normal force. The mass loss of rubber sample was measured by weighting the sample before and after each test to obtain the wear rate. The comparison between experimental and analytical results showed that the friction model could predict the friction coefficient accurately while the theory of powdery wear is unable to capture all the physics involved in rubber wear on rough surfaces.