The penetration of Ga along Al grain boundaries under stress-free conditions is investigated in the present study. In-situ SEM observations indicate that the penetration rate of Ga along Al grain boundaries at room temperature ranged from 6.4 to 9.2 μm/s, which is similar to the rate of diffusion in the liquid state. For a specific high energy grain boundary, the grain boundary misorientation is determined from the TEM diffraction Kikuchi pattern, and a molecular statics simulation method was employed to investigate grain boundary structure. A comparison of the structure of this high energy boundary is made with the Σ11(131)[101] tilt grain boundary that is not penetrated by Ga in the absence of the applied stress. The results indicate that the grain boundary plane void structure in the high energy grain boundary may provide void channels for Ga monolayer penetration. In addition, penetration behavior investigated under different length scales supports this model.

Molecular dynamics is used to study twinning deformation in a martensite microstructure obtained from rapid cooling β zirconium through the bcc-hcp transition. The microstructure is composed of (1011) twin boundaries and boundary dislocations which sometimes spread across the twins to form stacking faults. A series of such equilibrium microstructures subjected to discrete, increasing < 1012 > (1011) shear strain. The stress-strain curve has stick-slip behaviour with yield stress of ≈ 5.OKbar and yield strain of ≈ 3.8%. Deformation occurs through movement of twin boundaries in segments between boundary dislocations. Straight perfect twin boundaries do not move.

Commercial grain refiners for aluminium solidification are so potent that the barrier forgrain initiation is that for free growth rather than for nucleation itself. In this case quantitative prediction of grain size is possible. For small melt volumes a successful isothermal-melt model is presented. This is extended to directional solidification in a temperature gradient using cellular-automaton modelling of grain growth with finite-element heat-flow calculations.

A model has been developed which allows for the graphical representation of undercooling and nucleation site density in G-V space (i.e. temperature gradient G versus growth rate V). This model is used to explain a unique grain structure in weldments where equiaxed grains may be found both at the fusion boundary and along the center of the weld metal. Details of this model will be given and discussed in relation to grain refinement predictions in G- V space.

Existing experimental texture analysis capabilities allow testing of theories on plasticity using polycrystal models. Aluminum-lithium alloys, which are particularly suited for aerospace applications due to excellent strength to weight ratio, typically possess pronounced texture. Al- Cu-Li (2195) thick plates were deformed by cold rolling to various thickness reductions. The plates exhibit a texture gradient through the plate thickness accompanied by a variation in yield strength values. The difference in yield strength values is related to the texture variation in terms of the texture components (ideal crystallographic orientations) identified from experimental measurements. A modified Taylor-based polycrystal plasticity model developed by Kocks and his colleagues is used to predict yield surfaces by incorporating texture data. Results of this investigation show that the texture intensities measured at certain ideal orientations increase or decrease with increasing deformation. These textural changes influence the value and anisotropy of the yield strength of the alloy.

During the hot rolling of austenitic stainless steels, complete static recrystallisation is expected between passes unless finishing temperatures are low. Typically progressive refinement takes place to grain sizes in the range 20–50μm. However, most experimental studies of the effect of strain, strain rate, temperature and initial grain size on recrystallisation kinetics and recrystallised grain size under hot working conditions have been carried out on initial grain sizes greater than 50μm. Empirical relationships from these data and from more limited results of CMn steels have been extrapolated to smaller grain sizes for use in models of microstructural evolution during rolling.

Recent development of a physically based model for the effects of initial grain size, assuming that site saturated nucleation occurs at grain corners, grain edges, grain faces and at intragranular sites leads to interdependence of the effects of strain and grain sizeon nucleation density and hence on recrystallised grain size and recrystallisation rate. Experimental evidence available in the literature and some new results on finer grained Type 316 stainless steel are reviewed and compared with the expectations from the model.

The computer simulation of the evolution of microstructure during deformation processing is a desirable but elusive goal. In order to be effective, models must be tied to deformation parameters (temperature, strain rate, strain), and grain size distributions and recrystallisation kinetics must be predicted with high accuracy if the evolution of microstructure is to be tracked through several passes. This paper examines the cellular automaton (CA) simulation technique which has the potential to enable the modeller to accomplish these goals.

Aspects of the CA approach examined are the three dimensional representation of microstructure and its evolution and the incorporation of texture representation into simulations.Also investigated will be the limitations of the technique, in terms of the error that can be expected when different boundary conditions are imposed on a simulation.

Although at a rudimentary stage of investigation, the interfacing of CA simulations to deformation simulations (eg, finite element) will be examined. The paper concludes with a discussion of the challenges facing the implementation of CA simulations in industrial process models.

Plastic flow in metals occurs by a dislocation mechanism in which strain is accommodated by a dislocation flux. This results in tangled networks of dislocations which self-organize into stable cell structures with two characteristic length scales. These structures have a number of remarkable properties not least of which is that they exhibit scaling. We examine the idea that the evolution of geometrically necessary boundaries is dominated byrandom fluctuations, construct a simple model of cell orientations and show that scaling of misorientation distributions emerges a direct result of Gaussian white noise. We investigate whether hierachical microstructures containing both incidental dislocation boundaries and geometrically necessary boundaries can be studied using the same model. The results are compared with the experimental measurements and show good agreement.

Understanding the time-dependent flow behavior of metal alloys in semisolid state is essential for the further development of the process. In the present investigation, the thixotropic behavior of semisolid slurries is modeled using conservation equations and the Herschel-Bulkley fluid model. The rheological parameters are assumed to be functions of the solid volume fraction, and of a structural parameter that changes with processing history. The evolution of the structural parameter is described by a first order kinetic differential equation that relates the rate of build-up and break-down of the solid skeleton. The model is implemented into a computer code to predict die filling.

The precipitation kinetics of 7108.70 aluminum alloy has been investigated in a wide range of temperatures by in situ Small Angle X-ray Scattering (SAXS) and Transmission Electron Microscopy (TEM), and computer modeled by use of an internal-state variable model which predicts the evolution of microstructural parameters. The modeling and experiments were done for isothermal heat treatment at 120, 140, 150, 160 and 170°C. The industrial T6 and T7 treatments have also been investigated.