Peter J. Blau

Mechanisms for transitional friction and wear behavior of sliding metals

Abstract Four processes which may cause transitions in the unlubricated sliding behavior of metals are described. These processes are (1) metal transfer, (2) film formation and removal, (3) debris generation and (4) cyclic surface deterioration. Although these are not the only processes which may cause unexpected sudden friction and wear transitions, they may contribute significantly to many frequently observed sliding phenomena. It is possible that some combination of these mechanisms may act to effect changes in the wear and friction behavior of a given tribological system. The cyclic deterioration model involves four sequential stages in each cycle: (1) plastic deformation to obtain a highly deformed layer (HDL); (2) debris generation by crack initiation and growth in the HDL to produce flake-like debris; (3) debris removal through the shearing-off of load-bearing plateaus which slide along on their fracture surfaces to produce deep grooves; (4) plateau removal followed by a smoothing of the deep wear grooves (a process similar to the break-in of a rough surface with changes in the friction and the rate of wear). This process, when stages are occurring simultaneously on a large portion of the wear surface, could be observed as a cyclic variation in the friction and/or the wear rate.

Interpretations of the friction and wear break-in behavior of metals in sliding contact

Abstract Despite their complexities, friction and wear break-in behavior can provide important clues to the individual mechanistic contributions which interact to bring about longer-term sliding conditions. Because more than one mechanism could be responsible for given break-in curve shapes, more extensive experimentation in the effects of materials on the early stages of sliding needs to be undertaken. One method of relating friction and wear break-in behavior is the “break-in map” in which the duration ( i.e. time, number of cycles or sliding distance) required for the attainment of a constant average friction coefficient is simply plotted against the duration required for the achievement of a constant wear rate under the same conditions. A method of portraying frictional break-ins is to plot the difference in the friction between initial sliding and steady state sliding against the break-in duration. The friction-wear method may facilitate wear-in monitoring in machinery while the friction difference-break-in duration method may lead to a better understanding of sliding friction mechanisms.

A comparison of methods for determining wear volumes and surface parameters of spherically tipped sliders

Abstract The purpose of this paper is to investigate some of the methods of determining the wear volume and geometric surface parameters of a scar resulting from sliding either a sphere or spherically tipped pin on a nominally flat surface. There are several broad classes of methods for making this type of measurement. In this paper we will briefly describe these classes in general and then use abrasive wear data to illustrate three specific techniques in detail: two-dimensional analysis, done either manually or with an image analyzer; three-dimensional analysis where the X, Y, Z points came from a stylus-type profiling instrument; three-dimensional analysis where the X, Y, Z points were estimated from a two-dimensional projection obtained from a specially programmed image analyzer.

A comparison of four microindentation hardness test methods using copper, 52100 steel, and an amorphous PdCuSi alloy

Abstract Microindentation hardness (microhardness) numbers obtained with Knoop and Vickers diamond indenters can be greatly affected by indenter-test piece interactions. Traditional Knoop microhardness number equations assume that the ratio of the long diagonal to the short diagonal is constant for the impression. Data for large-grained pure Cu, fine-grained 52100 steel, and an amorphous PdCuSi alloy show that the diagonal ratio (defined in terms of a shape recovery index R) of relaxed impressions is seldom that expected for a Knoop indenter. The effects at several loads are examined. Four methods were used to obtain microhardness numbers: (1) a standard Knoop calculation (based on the length of the long diagonal), (2) a standard Vickers calculation (based on the average length of two diagonals), (3) a modified Knoop calculation of projected area hardness (based on the lengths of the long and short diagonals), and (4) an image analysis instrument measurement of the area of the impression (for the PdCuSi alloy only). A discussion of the results supports the accuracy of the two-diagonal modified Knoop method as a means to gain additional insight into the near-surface micromechanical properties of materials.

Investigation of the nature of micro-indentation hardness gradients below sliding contacts in five copper alloys worn against 52100 steel

This paper presents the results of a study of the variation of micro-indentation hardness with depth below sliding contact surfaces of OFHC Cu, Cu-3.5 wt % Al, Cu-7.0 wt % Al, and two commercial bronzes: CDA 638 and 688. All five metal alloys were worn dry against 52100 steel in a flat block (Cu alloy) on rotating cylinder (steel) configuration. The load was 10 N and sliding velocity was 20 cm sec−1 in a flowing argon environment. The variation of micro-indentation hardness with depth was found to be dependent upon the type of microstructural features below which each hardness profile was obtained. Therefore, micro-indentation hardness gradients sometimes varied more from location to location on a given sample than between similar microstructural features on one alloy and another. There was no obvious correlation between relative wear volumes of the alloys and the magnitude of their near surface micro-indentation hardness gradients. There did however seem to be a correlation between wear volumes and the thicknesses of highly deformed near-surface layers.

Effect of heat treatment and electron beam surface melting on the friction and wear behavior of a Cu-12wt.%Al alloy

Abstract A Cu-12wt.%Al eutectoid composition binary alloy was tested for friction and wear in three conditions designed to provide (1) a eutectoidal microstructure (treatment E), (2) a martensitic microstructure (treatment M) and (3) an electron-beam-melted surface (treatment EB). Polished blocks of the alloy were wear tested unlubricated against AISI 52100 steel rings at a 10 N load and 20 cm s −1 velocity in an argon gas environment. Both EB and M treatments gave lower block wear than the E heat treatment. All three showed transfer of material to the steel rings. Friction break-in characteristics varied with heat treatment. The martensitic microstructure, while lower in microindentation hardness, had a lower block wear volume. Electron beam melting of this alloy did not seem to improve performance any further than the quench used to produce martensite.

Use of a two-diagonal measurement method for reducing scatter in Knoop microhardness testing

Abstract Geometrical calculations and experimental measurements on Cu and Cu-30wt%Zn have demonstrated an inherent flaw in using conventional Knoop hardness values. Future studies could provide more meaningful microhardness-based information if the additional effort is made to use two-diagonal Knoop indentation measurement techniques.

Metallographic evidence for the nucleation of subsurface microcracks during unlubricated sliding of metals

Abstract The location, surface or subsurface, for nucleation and propagation of microcracks during sliding wear has been a continuing source of discussion and controversy in the tribology community. Rather than to suggest that there is only one correct answer to this question, the authors propose that, depending on the contact conditions and materials involved, several preferred sites for microcrack nucleation can exist. The present paper discusses a model which involves nucleation ahead of the slider, material flow under the slider, crack closure and reopening after the slider passes. Once microcracks nucleate, subsequent slider passes can cause crack propagation, coalescence and the eventual formation of wear debris particles. Recent metallographic evidence for crack nucleation ahead of a pyramidal slider which has a large negative rake angle is presented and suggests that this process is occurring to some extent; however, unequivocal substantiation of the models principal hypothesis still awaits research with a slider whose geometry promotes material flow underneath it rather than forming chips or flowing around the sides.

Initial frictional behavior during the wear of steel, aluminum and poly(methyl methacrylate) on abrasive papers

Abstract A series of experiments were conducted to investigate and model the friction break-in behavior for unlubricated reciprocating sliding of AISI 52100 steel, 2014-T4 aluminum, and poly(methyl methacrylate) on metallographic polishing papers. A ball-on-flat arrangement was used. Two representations of an analytical model were used, one of which had the capability to account for longer term friction transitions due to paper loading. Basic shapes of friction vs . cycle number curves were similar; however, the poly-(methyl methacrylate) and aluminum had larger changes during break-in and displayed longer term effects of paper loading.

An investigation of the unlubricated friction and wear break-in behavior of a dual-phase steel

Abstract The friction and wear break-in behavior of a dual-phase steel (steel DP 80) was investigated using a stroke-by-stroke fixed ball-on-flat tribometer. The standard rider balls were steel 52100 ball bearings slid at a load of 4.9 N and a velocity of 0.5 cm s −1 . The atmospheres used were air at a relative humidity of (58 ± 3)% and flowing argon gas at a relative humidity of 30% –35%. Surface finishes obtained by 1μm diamond lapping and by dry abrasion with 600 grit SiC cloth were tested. Frictional break-ins were shown to be reproducible on the SiC cloth samples and had a double peak in plots of the friction as a function of the number of strokes. Friction changes were less dramatic for diamond-polished flats. Argon environments reduced the friction and the wear track width. Freshly polished and cleaned surfaces broke in differently to those which had been stored in a desiccator for several weeks and then solvent cleaned before testing. The shapes of the friction break-in curves were altered in similar ways by changes in the surface finish and the test environment; this made it difficult to distinguish any unique effects of either of these variables.