A Model of Metal Fatigue

Abstract

THE important part played by the disturbance of the surface of a fatigue specimen by to-and-fro slip movement in the slip bands has been emphasized by much recent work, especially that of W. A. Wood¹. If the slip in alternate cycles were exactly reversed in the same atomic planes, then there would be no surface disturbance. On the other hand, extrusions or intrusions will be formed if by some mechanism such as that proposed by Mott² the dislocations which have moved in one direction along a certain path return by a different one, so as to go round a closed circuit in the same sense at each cycle. There is another possibility: that the dislocations make their return journeys along paths which are shifted in a random fashion with respect to the previous paths. If the shifts are comparable with the width of the slip band, this is equivalent to assuming that the slip is distributed at random at each half-cycle among the available slip planes, and that the distribution of slip at each half-cycle is independent of the distribution of slip in previous cycles.