Introduction to the discussion: the mechanism of friction

The application of the electron microscope, and of metallurgical, interferometric and other physical methods, shows that even carefully polished or cleaved surfaces have irregularities on them which are large compared with molecular dimensions. When two solids are placed together, the real area of contact is very small, so that the local pressure is high and, in general, exceeds the yield pressure, p m , of the metal. Plastic flow of the solid occurs at the summits of the irregularities so that the real area of contact A is proportional to the applied load W , i.e. A = W / p m . There is adhesion at these local regions of contact and the friction is, in a large measure, the force required to shear them. As a rough approximation F = As , where s is the shear strength of the junctions. Since A is proportional to W and independent of the size of the surfaces, this can explain the classical laws of friction. The mechanism of this process has been studied by optical and metallurgical methods and by the use of artificially radioactive metals. The strength and nature of the adhesion between metals is profoundly influenced by the oxide films which are normally present on them. It is the presence of these films which enables sliding to occur. If these surface films are removed in high vacuo and naked metals are placed in contact, there is strong adhesion—an attempt to slide them may cause further plastic deformation with consequent increase in the area of contact—and a large-scale ‘cold welding’ of the metals. If the sliding speed is appreciable, the temperature at the local points of rubbing contact will be raised. These surface temperatures have been measured by a thermo-electric method, by a visual or photographic method and by the use of an infra-red cell. All three methods give similar results and show that even at moderate speeds the local surface temperature may easily exceed 1000° C. It is limited by the melting of the metal. These high temperatures play an important part in polishing, the formation of the Beilby layer, the machining of metals and a number of other physical processes. Some low-friction materials are also described.

This paper describes an experimental study of the frictional behaviour of diamond, graphite and of carbon which have been outgassed in vacuo . The removal of surface films which are normally present causes a large increase in the friction. The admission of a small amount of oxygen, water vapour or other contaminant will reduce the friction. Both physical adsorption and chemical adsorption are important. There is evidence that with clean graphite surfaces there is strong adhesion at the interface, so that when sliding takes place slip and shearing occurs beneath the surface. Carbon and graphite have a negative tem perature coefficient of friction. The low friction normally observed with diamond is due to the presence of adsorbed oxygen and other gases. The friction of clean diamond on diamond is high, and the shear strength at the interface is comparable with the shear strength of diamond. Large-scale seizure does not occur because the deformation of the diamond in the region of contact is elastic and the real area of contact necessarily remains small.


1984 ◽  
Vol 106 (1) ◽  
pp. 26-34 ◽  
Author(s):  
Bharat Bhushan

The statistical analysis of the real area of contact proposed by Greenwood and Williamson is revisited. General and simplified equations for the mean asperity real area of contact, number of contacts, total real area of contact, and mean real pressure as a function of apparent pressure for the case of elastic junctions are presented. The critical value of the mean asperity pressure at which plastic flow starts when a polymer contacts a hard material is derived. Based on this, conditions of elastic and plastic junctions for polymers are defined by a “polymer” plasticity index, Ψp which depends on the complex modulus, Poisson’s ratio, yield strength, and surface topography. Calculations show that most dynamic contacts that occur in a computer-magnetic tape are elastic, and the predictions are supported by experimental evidence. Tape wear in computer applications is small and decreases Ψp by less than 10 percent. The theory presented here can also be applied to rigid and floppy disks.


Author(s):  
Yilei Zhang ◽  
Sriram Sundararajan

Autocorrelation Length (ACL) is a surface roughness parameter that provides spatial information of surface topography that is not included in amplitude parameters such as Root Mean Square roughness. This paper presents a statistical relation between ACL and the real area of contact, which is used to study the adhesive friction behavior of a rough surface. The influence of ACL on profile peak distribution is studied based on Whitehouse and Archard’s classical analysis, and their results are extended to compare profiles from different surfaces. With the knowledge of peak distribution, the real area of contact of a rough surface with a flat surface can be calculated using Hertzian contact mechanics. Numerical calculation shows that real area of contact increases with decreasing of ACL under the same normal load. Since adhesive friction force is proportional to real area of contact, this suggests that the adhesive friction behavior of a surface will be inversely proportional to its ACL. Results from microscale friction experiments on polished and etched silicon surfaces are presented to verify the analysis.


Sign in / Sign up

Export Citation Format

Share Document