The interpretation of certain phenomena occuring at nominally flat surfaces in stationary or sliding contact is dependent on the assumed distribution of the real area of contact between the surfaces. Since there is little direct evidence on which to base an estimate of this distribution, the approach used is to set up a simple model and compare the deduced theory (e.g., the deduced dependence of the experimental observables on the load) with the experimental evidence. The main conclusions are as follows. (a) The electrical contact resistance depends on the model used to represent the surfaces; the most realistic model is one in which increasing the load increases both the number and size of the contact areas. (b) In general, mechanical wear should also depend on the model. However, in wear experiments showing the simplest behavior, the wear rate is proportional to the load, and these results can be explained by assuming removal of lumps at contact areas formed by plastic deformation; moreover, this particular deduction is independent of the assumed model. This suggests that a basic assumption of previous theories, that increasing the load increases the number of contacts without affecting their average size, is redundant.

Contact and Rubbing of Flat Surfaces

The relation between the area of contact and the constriction resistance which holds for a single circular contact spot is widely used in electric contact theory, although the normal mode of contact is by a large number of microcontacts. A method of finding the resistance of a cluster of microcontacts is derived, and it is shown that the resistance may be regarded as the sum of the parallel resistance of the microcontacts and an interaction term often related to the extent of the cluster and not to the number or size of the individual contacts. The resistance is often close to that found by assuming that the entire area covered by the cluster is a single conducting spot. The known agreement between areas of contact found from resistance measurements and by other methods is therefore puzzling - until it is realized that the other methods also give only an apparent area: the real area of contact in, for example, a Brinell indentation is a small fraction of the area of the indentation. Thus from the point of view of electric contact theory the system is self-consistent, although the real area of contact is now seen to play no part in it: the implications for the theory of friction are more profound.

https://iopscience.iop.org/article/10.1088/0508-3443/17/12/310/pdf

Constriction resistance and the real area of contact

The implications of joint deformation in analyzing the properties and behavior of fractured rock masses, underground excavations, and faults

An apparatus for measuring the static coefficient of friction has been designed in which the forces tending to disturb the contact area are minimized and which further enables the contact area to be viewed throughout the experiment. Application of a tangential load, less than that required to give limiting friction, has been shown to give rise to a growth of both the real and apparent contact areas and to produce a relative displacement of the surfaces to an extent governed by the tangential load. An examination of the surface topography, by means of an electron microscope, has shown that the earlier type of slip ceases because surface scratches extend the boundaries of real contact Both the force of adhesion and the tangential force vary linearly with the real and apparent areas of contact. These results are in accord with a cohesive theory of friction, and offer no support for the theory on which friction is attributed to the formation of welded junctions.

http://iopscience.iop.org/article/10.1088/0370-1301/63/3/305/pdf

The Static Coefficient of Friction and the Area of Contact

The Cattaneo-Mindlin concept of interfacial slip in tangentially loaded compliant bodies is revisited and its basic simplifying assumptions are critically examined. It is shown that these assumptions, which, in the absence of modern numerical techniques, were essential in 1949 to enable an elegant quantitative solution of the basic problem of presliding between contacting bodies, are actually non physical. An alternative approach to the same problem that is based on treating sliding inception as a failure mode involving material plastic yield is discussed. This alternative approach was suggested even before 1949 but for the same lack of modern numerical techniques could only be promoted qualitatively. Some recent theoretical models, that are based on this earlier alternative approach, and in which the simplifying assumptions of the Cattaneo-Mindlin concept were completely relaxed are described along with their experimental verification. It is shown that the pre-sliding problem between contacting bodies can be accurately solved by these models using realistic physical assumptions and failure criterion.Copyright © 2009 by ASME

Revisiting the Cattaneo–Mindlin Concept of Interfacial Slip in Tangentially Loaded Compliant Bodies

Examination of the contours of the depression in a copper surface made by a cylindrical indenter shows that although considerable bulk deformation has occurred the asperities are relatively little deformed. When a similar groove is made by sliding with a hemispherical slider, there is sufficient surface damage to obliterate completely the surface irregularities. In the presence of a lubricant, similar effects are observed, though on a reduced scale. The tops of the asperities are wiped away, showing that there is considerable interaction between the metal surfaces through the lubricant film. A detailed examination of the surface damage produced during sliding shows that metallic junctions are formed and sheared during the sliding process. These junctions are formed even when the sliding speeds are so small that the temperature rise due to frictional heating is negligible, and it is suggested that they are produced by a cold welding of the surfaces as a result of the high localized pressures developed at the points of real contact. These junctions are often strong enough to rupture the stronger of the two sliding metals. Thus when copper slides on steel (unlubricated) there is not only a marked tearing out of copper fragments from the copper surface, but also a plucking of minute particles of steel out of the steel surface. Further, the shearing of metallic junctions and the smearing of metallic fragments over the surfaces may produce considerable work hardening below the actual points of contact. Although a lubricant reduces the amount of intimate metallic contact, the investigation shows that metallic junctions are formed through the lubricant film by an essentially similar mechanism. These observations provide graphic evidence for the view that the frictional force for both clean and lubricated surfaces is due mainly to the shearing of metallic junctions formed by cold welding at the points of intimate contact. The significance of these conclusions in the burnishing of metals and in the behaviuor of bearing alloys is discussed.

A. J. W. Moore

Papers

Deformation of metals in static and in sliding contact