Size effect, internal stress, and diffusion driven boundary motion
- PDF / 1,024,152 Bytes
- 4 Pages / 603.28 x 788 pts Page_size
- 109 Downloads / 179 Views
I.
INTRODUCTION
DIFFUSION induced boundary motion in metals at low temperature has now been reported for many systems. 1-9 In addition to the movement of existing boundaries new grains may nucleate and grow under this driving force (free energy of mixing). While this type of boundary motion has been observed in many systems, it is not observed in other systems which exhibit the required solubility and thus the required driving force (for example, Ag on Cul~ Also, though the process is driven by a force greater than any restraint due to boundary curvature or inclusions, the boundaries often stop and reverse after transforming only a limited part of the sample surface. Furthermore, there has been little work on how deep and how fast this DDBM transformed region will penetrate into a thicker sample. A different set of questions revolves about the topic of what kind of boundaries can give rise to this effect. For example, if recrystallization occurs in the presence of such a driving force, will DDBM occur? Or, if strain induced boundary motion occurs in a system where DDBM is not normally observed, will DDBM be induced or accompany the boundary motion? Grain boundary diffusion is necessary for DDBM, but is it sufficient? The only force large enough to stop DDBM, that has been suggested to date, is that of long range elastic stress induced by the volume change accompanying the introduction of solute into the sample through DDBM. Shewmon has shown that there will be a characteristic shape change accompanying DDBM which reflects the relative grain boundary diffusivities of the two species. For example, if the grain boundary diffusion coefficient of Zn, D'(Zn), was much greater than that of Fe, D'(Fe), in the DDBM driven boundary, the addition of solute will expand the region in the plane of the sheet (normal to the moving boundary) to accommodate the added atoms. However, if D'(Zn),~ D' (Fe) the large flux of Fe out of the sample, as Zn diffuses P.G. SHEWMON is Professor and Chairman, Department of Metallurgical Engineering, The Ohio State University, Columbus, OH 43210. ZONG-SEN YU, formerly at The Ohio State University, Columbus, has now returned to Peking Institute of Iron and Steel Technology, Peking, People's Republic of China. Manuscript submitted November 29, 1982. METALLURGICAL TRANSACTIONS A
in, would create a surface step at the advancing interface with the alloy grain higher than the surface of the pure iron grain. The buckling and wrinkling of 37/zm thick foil of pure polycrystalline Fe undergoing DDBM 6 was interpreted to indicate that D'(Zn) ~ D'(Fe). If such buckling occurred in thin sheet when the boundaries moved through the entire thickness, roughly as a plane, what would happen in thick samples? Since grain boundary diffusion can move solute in to only a finite depth, coherence between this surface layer and the bulk of the sample would set up large stresses that would severely restrain the development of a DDBM layer. This work was undertaken to study the effect of specimen thickness and strain indu
Data Loading...
