On Strain-Enhanced Diffusion in Metals. III. Interpretation of Recent Experiments

Abstract

A critical interpretation of recent experimental results on the effects of plastic deformation on diffusion is carried out with the aid of the results established in the preceding Parts I and II. Attention is focused on experiments carried out at elevated temperatures (higher than half the absolute melting temperature) in which diffusion occurred over macroscopic distances. The results are discussed in terms of the following possible causes of apparent enhanced diffusion: (1) generation and migration of excess point defects during deformation; (2) short‐circuiting along static or moving dislocations or grain boundaries created in the deformation; (3) short‐circuiting along cracks created in the deformation; and (4) experimental difficulties associated with interface roughness. Causes (3) and (4) are of little basic interest but must be considered in any survey of experimental results. Fourteen experiments in all are considered. Ten of these experiments showed no apparent enhancements within the expected experimental error of about 50% in the diffusivity. In one of these ten experiments a large enhancement was reported, but is is shown that this was due solely to an error in mathematical analysis. Two of the experiments showed large apparent enhancements, and it is shown that they were almost certainly due to (4). One of the experiments showed apparent enhancements due to (3) but showed no enhancement in the absence of cracking. One of the experiments showed enhancements which, it is concluded, are not inconsistent with present knowledge of (2). [Very recent work by Brown and Blackburn (unpublished) indicate that at least some of the enhancements in this experiment may have been due to (4)]. It is generally concluded that in no case has a diffusion enhancement been detected due to the generation of excess vacancies. However, enhancement in one case may have occurred as a result of short‐circuiting. These results are consistent with other high temperature deformation phenomena, e.g., creep kinetics, and are also consistent with the point defect and short‐circuiting models developed in the preceding Parts I and II which predict very small effects due to point defects and possibly larger effects in certain cases due to short‐circuiting.