S. Saimoto

Study of large strain deformation of dilute solid solutions of Al-Cu using channel-die compression

Abstract The mechanical properties of pure aluminium, Al-0.2 wt.% Cu and Al-0.4 wt.% Cu at large strains were studied using channel-die compression at three different temperatures: 77, 200 and 295 K. The evolution of the structure was studied by various techniques, including slip line studies, global texture measurements, back scattered electron Kikuchi patterns and TEM. Intense shear banding was observed at large strains and is related to the texture evolution. The mechanical behaviour was interpreted in terms of work hardening rate vs. stress plots (or θ/τ plots). Increasing the temperature leads to a decrease of the initial slope of the θ/τ plot as well as an increase in its concavity. At large strains all materials exhibited a stage of constant work hardening at low rate, or stage IV. The addition of solute was shown to result in an increase of the work hardening rate, which could be represented by a translation of the θ/τ plot on the stress axis. Phenomenological models are proposed for the prediction of the influence of temperature and solute content on work hardening.

Dynamic dislocation–defect analysis

The dynamic internal variables which control plastic flow can only be assessed by dynamic materials testing at any given instance. The testing method championned by our studies has been precision strain rate sensitivity (PSRS) whereby the change in flow stress due to a set change in strain rate is taken to be an operational measure of the activation volume and its product with the flow stress gives rise to the operational activation work. Also, from the work hardening slope, a modelled parameter proportional to the mean slip distance (λ) is simultaneously determined. The deviation from the linear Cottrell–Stokes relation as determined with the Haasen plot indicates the evolution of secondary defects other than monopole dislocations. Hence PSRS can assess the theoretical predictions of the activation distance (d) and work as a function of temperature, resulting in quantitative values that are in accord with dislocation theory at temperatures below that where point defects become mobile. A method to calibrate λ using Stage II slope θII shows that λ/ℓ, where ℓ is the mean forest dislocation spacing, is inversely proportional to θ, the work hardening coefficient. This analysis has led to a new plot of θII/θ versus b 2λ/ν where b is the Burgers vector and its slope is directly proportional to d. An example using an alumina-dispersed high conductivity copper shows that geometrically necessary punched out loops are continuously generated. The role of point defect mobility is dramatically illustrated by load drops in [001] aluminium crystals with the formation of slip clusters.

The determination of solute clusters in dilute aluminum alloys using strain rate sensitivity

Abstract Precise, apparent activation volumes, V′, are determined at 78 K for aluminum solvent systems containing mostly Fe, Cr, Cu or Mg in solution. The relationship between 1/V′ and the flow stress correlates to the amount of solute in the aluminum matrix. It appears that only Fe solute manifests solute-defect interaction effects which approximate a random solid solution. All other solute elements exhibit a much reduced distribution of strain centres, suggesting that these solutes are clustered even though the dilute alloys were solutionized at temperatures considerably above the solvus temperature. Stress equivalence of the apparent activation volume is observed over a range of yield stress, σ0, as V′ ∝ σ0−0.95.

Residual stresses in discontinuous metal matrix composites

Abstract Residual microstress was measured, using X-ray and neutron diffraction, in SiCAl composites with SiC in the form of whiskers, platelets and spheres. Average whisker and sphere stresses are similar, with platelet values considerably lower, in disagreement with finite element method analyses. Reasons for the discrepancy, and a comparison of the X-ray and neutron results, are discussed.

Dislocation-defect analysis during micro-indentation using displacement rate changes

Abstract Micro-indentation with step-ramp displacement rate changes has been developed to nullify the machine and specimen compliance and to directly measure the apparent activation work, Δ w′, of the deforming material. The Cottrell-Stokes relation is obeyed for an eutectic solder, but in contrast, nominally pure Al in the annealed and pre-rolled states indicate that the type of obstacle barrier continuously changes as a function of strain rate. Thus it is deduced that for high homologous temperature testing, either dislocation-dislocation interactions or jog dragging are detected, whereas at lower temperatures other defects are generated especially when the total plastic volume under the indenter is minute. These results can be correlated to the evolution of debris at high strains detected by tensile tests.

Point defect generation, nano-void formation and growth. II. Criterion for ductile failure

Using the derived relation for point defect generation according to a new constitutive relation, the notion of nano-void formation at grown-in nano-particles is examined and its consequences deduced as the nano-voids grow in size with continued deformation. Assuming that void growth is due only to point defect accumulation, the analysis of fracture strains in tension of natural-aged AA6111 suggests that coalescence by micro-plastic activity occurs when the void diameter becomes about one third of the evolving inter-void spacing. Hence, the derived limit strain to incipient void-coalescence is inversely proportional to the square root of point defect generation as determined from the stress–strain data. Using this criterion, failure prediction maps can be constructed for strain modes of plane-strain and balanced bi-axial to result in the outer bounds of the forming limit diagram. Trial examinations with AA5754 and AA3003 show great promise.

Point defect generation, nano-void formation and growth. I. Validation

The volume fraction of point defects generated as a function of plastic shear strain squared, γ2, was derived from crystal plasticity concepts. The evolution was determined from the stress–strain values using a new constitutive relation which replicates the measured behavior with at least two fitted loci. Assuming that nano-voids form by clustering of vacancies, the nano-void diameter was found to be proportional to their spacing and shear strain with the constant being characteristic of point defect production during deformation. The predicted amount of point defect generated was validated using the previously determined resistivity of [100] copper single crystals deformed at 4.2 K and annealed at 296 K. Similar analysis of super-pure polycrystalline copper data affirmed that the dynamic annihilation parameter extrinsically incorporated in the new derivation is larger due to formation of slip clusters. Moreover, the temperature dependence of the mean slip-distance to inter-forest spacing ratio at Stage II to III transition indicates that the thermally activated drag of vacancy-creating jogs occurs above 150 K. For polycrystalline aluminum deformed at 296 K, it was concluded that the nuclei of the nano-voids were not part of the evolving dislocation array but were embedded in the grown-in microstructure. This hypothesis is pursued in the accompanying paper, Part II, and its prediction results in a criterion for ductile failure.

Kinetic analysis of dynamic point defect pinning in aluminium initiated by strain rate changes

The role of point defect production during deformation was examined by sealing the vacancy sinks in the grain boundaries with solutes to magnify its effect upon instantaneous strain-rate changes. AA1100 aluminium sheets were thermal-mechanically treated to result in a grain size of about 25 µm and in grain boundaries that were not capable of acting as efficient vacancy sinks. Tensile tests at various temperatures ranging from 78 to 300 K showed that above 195 K, the pinning effect could be quantitatively analysed. A rate equation analysis for mono- and di-vacancy recovery was adopted to perform fits to the deduced change in flow stress with time after strain-rate change from which apparent activation energies were derived. This examination indicates that the migrating species are predominantly di-vacancies. It is concluded that point-defect atmospheres have the capacity to glide in unison with mobile dislocations and hence are sensitive to the magnitude of the strain rate and temperature.

Effects of Solubility Limit and the Presence of Ultra-Fine Al6Fe on the Kinetics of Grain Growth in Dilute Al-Fe Alloys

A nominally pure Al slab was thermo-mechanically treated to result in a near random texture of 90 m grain size. Subsequent cold rolling with intermediate anneals at 230, 275, and 300°C reduced the Fe solute to near equilibrium compositions below 0.5 ppm atomic. The final cold rolled sheet continuously recrystallized; grain growth of this structure is reported. A grain-growth kinetics mapping was generated, correlating the parameters of Fe-in-Al solubility limit, Fe diffusivities in the grain boundaries and the Al lattice and the activation energies for migration rates.

A unifying analysis of the microplastic kinetics with the operative plastic potential

The studies on the existence of a plastic potential in metal systems pre-dates by decades those of thermally activated flow as elucidated by the seminal work of Basinski in 1959. However, to this date a direct correlation between the microplastic kinetic events which integrally manifests itself as the macroscopically determinable plastic potential has not been made. In this work a trial notion of using the activation volume as the fundamental internal variable results in re-expressing the macroscopic rate of doing work in terms of the work of activation necessary to overcome an internal discrete obstacle to dislocation passage. The subsequent differentiation results in a rate of power dissipation relation which under strict conditions of constant-strain-rate sensitivity is only dependent on the square of the strain rate and a linear work-hardening coefficient. On the other hand, if the time dependences of the internal parameters are considered, the power dissipation rate becomes a function of not only the strain rate but also the microstructural evolutionary parameters. The examinations of these predictions suggest means of materials characterization to assess which alloy heat treatment will enhance greater ductility.The studies on the existence of a plastic potential in metal systems pre-dates by decades those of thermally activated flow as elucidated by the seminal work of Basinski in 1959. However, to this date a direct correlation between the microplastic kinetic events which integrally manifests itself as the macroscopically determinable plastic potential has not been made. In this work a trial notion of using the activation volume as the fundamental internal variable results in re-expressing the macroscopic rate of doing work in terms of the work of activation necessary to overcome an internal discrete obstacle to dislocation passage. The subsequent differentiation results in a rate of power dissipation relation which under strict conditions of constant-strain-rate sensitivity is only dependent on the square of the strain rate and a linear work-hardening coefficient. On the other hand, if the time dependences of the internal parameters are considered, the power dissipation rate becomes a function of not only the...