Effect of mechanical properties on the adhesive wear of metals

Abstract

There have been consistent efforts to understand the wear behaviour of materials. Many of the investigations were carried out with the aim of correlating wear behaviour with the basic properties of the material in order to determine the property that most controls the wear process. Since wear is a complex phenomenon, not yet fully understood, and many types of wear exist, such correlations must be specific in giving the particular test conditions. Since two materials are involved in wear, and the effects are confined to regions very close to the interface, the correlations must in some way consider the relative surface properties. Many investigators [1-3] have been successful in relating wear resistance, especially to abrasive type of wear, to the hardness of the materials to show a nearly linear variation. Correlations with other properties have been limited in extent. Theories of wear were proposed a long time ago by Finkin [4], Merchant [5], Feng [6], Bowden and Tabor [7] and Kerridge [8], and have been modified to apply to many wear situations. Often results seem to indicate [9] that wear depends only on the conditions bearing load, sliding speed and relative hardness of the softer surface. Evidence has also been obtained [5, 7] to show that it is more the yield stress of the softer surface than any other property that controls wear. Burwell [2] proposed a model for calculating the wear rate of metals, in which the relative hardness was more important than any other property, although data obtained on the basis of this model were highly scattered. The model of Archard [9] was not applied to the case of pure metals. The most significant correlation seems to be that of Rabinowicz [10], in which wear was shown to be related to the surface energy. It is often believed that the friction at the wear interface is closely related to the wear behaviour. However , wear situations can exist in which the wear rate can vary abruptly with no significant change in the friction coefficient. However , a recent study has [11, 12] shown that the coefficient of friction measured in a pin-on-disc machine for a number of pure metals bears a predictable correlation with several mechanical properties of the metal. However , correlations do not seem to have been made for metals for their wear behaviour under purely adhesive conditions. It has been recognized [13] that nearly all materials tend to form a wear-resistant surface layer under adhesive conditions, to justify the comment that wear resistance is more a function of the properties of the surface layer than that of the bulk. It was therefore thought that correlating the properties of the metal with the adhesive wear behaviour would yield interesting results, and this was attempted. Part of the results obtained are summarized in this letter. Cylindrical wear pins 5 mm in diameter were obtained from rods of the desired metals (Table I). One flat end of the pin was ground with 600-grade emery and then slid on a rotating highly polished steel surface of a hardness 50 H R C machine for a total of 30 min. The bearing load used was 5 N and the relative sliding speed at the wear interface was 1.25 m s -1. Steady-state wear rates were calculated from the weight measurements and expressed as volume worn per unit sliding distance (cm 3 cm-1). The measured wear rates in all of the metals studied for the conditions of the experiment were-of the order of 10-9-10 .8 cm 3 cm -1 and are considered