佐藤 英一

A new theoretical model is proposed to explain internal stress superplasticity under a simultaneous applied stress during thermal cycling. The analyzed material is an elastically uniform body containing an elastic spherical inclusion, surrounded by a plastic matrix obeying a power law creep. It is assumed that only relaxation by interface diffusion is significant. At first it is shown that an inclusion dilatation can be counterbalanced by a certain matrix plastic flow. Assuming a certain stress distribution for this condition, a stationary flow on heating or cooling results. Typical behavior of this stationary flow is analyzed for the specific cases of high or low applied stresses. Under a low applied stress, the internal stress by an inclusion dilatation strongly accelerates the flow in the direction of the applied stress, which is in proportion to the applied stress, to the heating or cooling rate and to the inclusion volume fraction.

In the transparent organic crystal known as succinonitrile-acetone binary alloy, transient behavior of unidirectional solidification is directly observed within the range where the planar interface is stable, by means of a microscopic interferometer. Interface morphology and solidification rates are obtained by bright-field observations. Interference fringes are used to determine the gradient of the temperature and of the solute concentration in the liquid ahead of the solid-liquid interface. Although the solidification direction is taken such that the thermal convection is suppressed, experimental data on solidification rates and concentration gradients agree well with numerical values based not on the diffusion-controlled model, but on the boundary layer model which assumes fluid mixing beyond the boundary layer. One of the reasons why fluid mixing occurs is thought to be the thermosolutal convection induced by the concentration gradient built up ahead of the solid-liquid interface. Thickness of the boundary layer estimated from experimental data of solidification rates agrees quantitatively with those obtained from interference fringes and from tracer analysis.

Superplastic deformation, especially in quasi-stable fine equiaxied grain structures, is accompanied by grain growth whose rate exceeds that which can occur without deformation. The deformation induced component of the grain growth stabilizes the deformation itself through an increase in the flow stress. An empirical expression for the grain growth is demonstrated which describes the behavior of microduplex and second-phase dispersed alloys. Finally, a new deformation model of superplasticity is proposed to explain the grain growth.

Deformation induced grain growth is usually observed during structural superplasticity. This paper proposes a new deformation model explaining this grain growth based on grain switching proposed by Ashby and Verrall. To begin with, an irregularity of a pentagon-heptagon pair is introduced into a two-dimensional grain structure constructed with a regular hexagonal array. This irregularity geometrically corresponds to an edge dislocation of the grain structure. The tensile stress oblique to the pair results the grain switching described by dislocation glide. On the other hand, the stress parallel to it results the grain switching which transforms the pentagon into the quadrilateral and then removes it. It can be described by dislocation climb and results the enhancement of grain growth. Therefore superplastic deformation (grain switching) induces grain growth due to the strain. In order to extend this model to three dimensions, an edge dislocation is introduced into a three-dimensional grain structure constructed with Kelvin's tetra-kai-dekahedra (14-faced grains). The grain at the tip of the half grain plane is 13-faced or 11-faced. A 17-faced or 15-faced grain is contacting it, respectively. Since these coincide with the pentagon-hexagon pair in two dimensions, the above discussion in two dimensions can use in three dimensions with little modifications. This model predicts the grain growth rate equation as [GRAPHICS] This well agrees with the experimental results in dual phase structure.

The electronic structure of (TaSe4)2I has been studied by positron annihilation experiment and ultraviolet photoemission spectroscopy (UPS). By positron study, the Fermi surface is observed only for the high conducting c-axis direction, elucidating the one-dimensionality of the electronic structure. The Fermi wave vector kF is also determined. By UPS study, it is found that the hybridization of Ta 5d orbitals with Se 4p orbitals is remarkable at shallow binding energies in the valence bands. © 1986.