On Strain-Enhanced Diffusion in Metals. I. Point Defect Models

Publisher Website
Google Scholar
Abstract
Plastic deformation may affect bulk diffusion by altering the rates of point defect production and annihilation. In the present work (Part I) a detailed analysis of this phenomenon is given with particular emphasis upon possible upper limits for the magnitude of the effect at elevated diffusion temperatures where bulk diffusion occurs over macroscopic distances. In order to consider all possibilities from this point of view the following three models are analyzed: model I—vacancies may be created or destroyed at jogs on gliding screw dislocations and climbing edge dislocations. These possible vacancy sources and sinks are intermixed within crystal subgrains and the net source or sink action is zero over representative subgrain volumes; model II—a net vacancy source (or sink) action is required within the subgrains in order to support the required nonconservative dislocation motion. The required excess vacancies are destroyed (created) at the subgrain boundaries where the concentration is maintained in equilibrium; model III—large numbers of vacancies are readily created everywhere during deformation but are destroyed everywhere in the specimen volume only with difficulty. It is concluded that model I is probably the most realistic. In this case, upper limits are obtained for the required defect production and destruction at dislocations, and it is found that the required defects can be obtained from rather small perturbations of the usual thermally generated fluxes in detailed balance at the jogs. At practical diffusion temperatures and strain rates of the effects of the deformation on bulk diffusion are, therefore, very small and are below the limits of detection by usual methods. Model II appears to be conceivable under certain conditions. Analysis of this case indicates that the effects of deformation are larger than in model I because of the longer range vacancy diffusion which is required. However, the effects are still small under usual conditions. Model III is purely phenomenological and is included since it has been used by others. Several characteristics of this model which may be used to set limitations on possible enhancements obtainable in this model are derived. In addition to the analysis of the above three models, detailed discussion is given of the following aspects of the diffusion—deformation problem: (1) effect of surface vacancy sinks on diffusion—deformation experiments carried out near a free surface; (2) diffusion enhancement expressed in terms of the number of excess vacancy jumps; (3) temperature dependence of the diffusional effects of deformation; (4) mechanical work required to maintain nonequilibrium vacancy concentrations; (5) the relationship between macroscopic creep behavior and nonequilibrium vacancies; (6) the possibilities of interstitial production. Many of the results established here are employed in Part III [J. Appl. Phys. (to be published)] where a comprehensive interpretation of recent experimental work in this field is carried out. Part II [J. Appl. Phys. (to be published)] consists of an analysis of strain-enhanced diffusion due to short-circuiting along static and moving dislocations and grain boundaries.