The geometric configuration of grain boundaries and surfaces seems to play a significant role in phase transformations and surface phenomena. Determining the equilibrium configurations of such boundaries for a given surface tension function γ is additionally an interesting mathematical problem; it reduces in the case of isotropic surface tension to the minimal surface problem. A framework is given here for determining such configurations in the other extreme case, where the equilibrium shape of a crystal of fixed volume (the Wulff shape) is a polyhedron. Some results obtained within this framework are outlined.

The influence of surface magnetization on the surface segregation in binary alloys with one magnetic component is studied. Based in a model in which the local magnetic moment on the ferromagnetic atoms is determined by the local environment, the internal energy is calculated within the mean field approximation. We applied the model to dilute CuNi alloys and we find that magnetism favors the segregation of Ni, contrary to what it is found in the paramagnetic phase.

The power laser has been applied to the surface melting of a high-carbon alloy steel with laser operating parameters leading to structures which cannot be produced by the usual heat-treatments. The subsequent annealing of these structures has also been investigated. Optical and scanning electron microscopy, X-ray diffraction and Mössbauer spectroscopy have been used as methods of analysis.

Graphite intercalation compounds offer a unique laboratory in which to study many kinds of phase transitions. These include melting, reorientations, staging, commensurate-incommensurate effects, and certain types of lattice gas transitions. Molecular and alkali-metal intercalates are discussed with reference to results of X-ray diffuse scattering studies as a function of temperature and pressure.

Electronic excess energy of a composition-modulated alloy system at absolute zero temperature is obtained using Hohenberg and Kohn's formula (for the ground-state energy of an inhomogeneous electron gas) in a modulated positive-background model. Dependence of the electronic excess energy on the modulation wavelength is studied. Two leading terms in the excess energy are examined to elucidate the limitation of this model.

Artificial superlattices are multilayer structures consisting of a onedimensional sequence of ultrathin layers of different semiconductors (compositional or hetero junction superlattices) or layers of the same material, modulated by periodic n- and p-doping (doping or p-n junction superlattices). The electronic properties of these new materials differ qualitatively from those of their components and can be tailored within wide limits by appropriate choice of the design parameters (i.e. superlattice period, alloy composition, and/or doping concentration). Different electronic phases (insulating, semiconducting, semimetallic, metallic, e.g.) may be obtained in this way.

Molecular dynamic computer simulations have recently provided some very graphic portrayals of how model lattices can surmount nucleation barriers of various types and accomplish certain kinds of first order displacive transitions. A survey of the different two-and three-dimensional transformations observed to date, along with relevant details about inter-atomic potential, potencies of nucleating defects and preferred transformationpaths will be presented. The simulation of other types of first order transformation in solids, such as melting and allotropic transformations will be touched upon. Methods by which a close representation of martensitic transformations in real systems should be obtainable will be discussed.

Investigations of the shape memory effect in alloys forming thermoelastic martensites with various crystal structures (2H, 3R, 9R and 18R) reveal that universal behavior exists. A unified explanation of the martensite deformation processes and subsequent shape recovery is now in hand, even though the various martensites are both internally twinned and internally faulted and, in addition, have different crystal structures. In cases studied to date, an initial parent phase single crystal transforms into self-accommodating arrangements of martensite variants (plates) which are characterized by “plate groups.” Each group consists of four variants. The average shape deformation in a plate group is essentially zero.

Upon stressing below the Mf temperature the martensite undergoes deformation by detwinning (2H and 3R only), variant-variant coalescence and twinning processes, and further group-to-group coalescence. The deformed specimen eventually becomes a single crystal of martensite consisting of that particular habit plane variant whose shape deformation permits maximum extension in the direction of the applied stress. The deformed martensite persists after unloading has occurred; reverse rearrangements of twins and variants do not occur. Specimens deformed below Mf regain their initial shape characteristic of the initial parent phase upon heating from As to Af, during which the single crystal of martensite obtained by stressing the 24-variant configuration transforms back to the original parent phase single crystal in a unique manner, which is basically a simple “unshearing” process. The unshearing is the essence of the memory.

The two-way shape memory effect results after the initial martensitic transformation upon cooling is preprogrammed by the introduction of stresses which preferentially bias the transformation so that only a single variant of martensite forms upon cooling. The shape change of this single variant causes the characteristic spontaneous “bending” upon cooling. The characteristic “unbending upon heating is as with the conventional “one-way” shape memory effect.

The character and mechanism of two-way shape memory (TWSM) “training” has been investigated in a Cu-Zn-Al alloy by means of detailed thermomechanical evaluation and TEM observations. The progressive effects of training on the TWSM behaviour, as well as an accompanying substructural evolution, have been established. The results indicate a relationship between the substructural effects of cyclic thermoelastic martensitic transformation and the ability to exhibit TWSM. Microscopic studies reveal that as the number of cycles of thermoelastic transformation under stress increases, specific physical features develop in the parent phase submicrostructure. These take the form of dislocation structures which evolve into “vestigial” martensite markinqs. These in turn assist in the nucleation and growth of a preferred martensite plate arrangement on cooling which is similar to that induced under stress during the training cycles.

The lattice parameter and Curie temperature of austenite in Fe3Pt increase sharply on ordering for times up to 2 hr at 650° C. These effects, together with a rapid decrease in the martensitic transformation temperatures, are interpreted in terms of increased short-range order. The onset of thermoelastic reversibility of the martensitic transformation is associated with the development of a significant volume fraction of long-range ordered domains.

An unexpected 18R martensite to martensite transformation was obtained in Cu-Zn-Al alloy during the training process for 18R single crystal preparation: this later was induced from a β1 single crystal by application of a constant tensile force at room temperature and by lowering the temperature until β1 -->18R transformation occurs. At this condition, and after several cycles, some plates of a new phase were observed to appear in a sudden way inside the 18R single crystal region. By studying the transformation characteristics as given by elongation temperature curves and trace analysis we conclude that this new phase plates are 2H martensite.

We demonstrate unambiguously that V5Si3 inclusions coherent with the V3Si matrix induce at low temperature a tetragonal distorsion of V3Si.

a) Polycrystalline samples are transforming only if they are two phased V3Si - V5Si3,

b) the oriented eutectic V3Si - V5Si3 undergoes a cubic - tetragonal distortion under 60K.

c) A transforming V3Si crystal is obtained by solid state epitaxy of a V5Si3 layer on the (001) plane.

These experiments support very well our idea that the distorsion can be driven by internal stresses in the samples, related to silicon over stoichiometry.

The aim of this paper is to study the contribution afforded by strain induced martensitic transformation to the plastic behavior under monotonic or cyclic loadings and the mechanical properties of Cr-Ni austenitic stainless steels. The study concerns namely the low carbon 18/10 steel, commonly used in various fields of technics and particularly in cryogenics (liquid gases storage or transportation).

A new lattice correspondence has been deduced for the γ→a′ martensite reaction in dilute uranium alloys. The new correspondence involves a lattice deformation with significantly smaller principal strains than those previously considered, although more atomic shuffles are required.

From an analysis of possible transformation twinning modes based on the new correspondence, the further possibility of a deformation twinning mode in these alloys is theoretically predicted. The mode is compound, described by (021)(023) [012] [032], and involves a twinning shear (g = .044) which is smaller than any value previously predicted or observed, although again a larger number of shuffles would be required. The significance of these results is discussed.

This paper is intended to describe and illustrate the use of phase transformations in the design of new alloy steels. The general method of alloy design is described. Two examples are discussed: the development of new ferritic structural steels for use at cryogenic temperature, and the development of ‘dual phase’ steels for automotive use.

In AlMgSi-alloys cooled from about 773K (500°C) the strength after ageing depends on the rate of cooling. By addition of Mn or Cr dispersoids are formed which may control grain structure. The dispersoids also act as nucleation sites for coarse β′-Mg2Si particles. This makes the alloys more quench sensitive, in this paper the influence of 0.56%Mn on quench sensitivity is compared to that of 0.23%Cr in Al–O.6%Mg-1.0%Si industrial alloys.

The alloys were homogenized at 793K (520°C) and 853K (580°C), extruded, quenched in water or cooled in air, and aged to maximum hardness. The density of dispersoids varied with homogenizing heat treatment. Quench sensitivity increased with increasing density of dispersoids.

The two alloys had about equal densities of dispersoids after homogenizing at 793K (520°C), but the alloy containing Cr was found to be more quench sensitive than the alloy containing Mn. This is discussed and compared with other findings.

The phase transformations occurring on quenching and subsequent aging of a Cu-Al-Ni β phase alloy have been studied using transmission electron microscopy, x-ray diffraction and optical microscopy. Quenching produces an ordered solid solution, β1, based on the DO3 structure of Cu3A1. X-ray and electron diffraction show satellite diffraction peaks. On aging, these satellite peaks disappear and the parent phase undergoes a martensitic transformation. It is suggested that the parent phase undergoes a process of phase separation through the mechanism of spinodal decomposition before the martensitic transformation occurs. The martensitic transformation shows isothermal kinetics for which a transformation diagram is presented. Prolonged aging results in the formation of equilibrium phases.

The Charpy ν-notch impact properties at -196°C of casts of Type 316 austenitic stainless steel have been compared, after ageing at 650°C for time periods up to 90 minutes. The impact energy of all steels depends on ageing time; the degree of response to ageing depends on original delta ferrite content, which varied from −0.5% to 5.0% in the steels studied. For steels with delta ferrite content in the lower half of this range the impact energy, during ageing, initially decreases to a minimum, subsequently rises to a maximum and finally decreases. Steels with delta ferrite contents near 5% suffer a continuous decrease in impact energy as ageing time increases.

Microstructural investigation reveals, during early stages of ageing, the rapid formation of sigma phase, at the ferrite-austenite interphase boundaries. Some delta concurrently transforms to new austenite. Microfractographs indicate a change in the nature of the fracture as ageing progresses. Initially, fractures have large dimples, but as ageing progresses, a finer dimpling appears and extends in area. These dimples contain fine particles, identified mainly as MnS. The variations in impact behaviour are explained on the basis of the structural changes observed.

It is general practice to weld bullet proof steels with 20Cr-9Ni-3Mo Duplex stainless alloy. Little information is available regarding increasing the strength of these weldments or their response to time at temperatures used for heat treatments. This investigation sought to study possible strengthening mechanisms involving silicon additions and the effects of heat treating such welds at 1050°C.

A 2% silicon addition gave a 16% increase in strength with approximately equal contributions from stacking fault energy, solute and second phase hardening. However, the ductility was significantly reduced by a high ferrite content. When this δ ferrite content was corrected by cobalt additions a 26% increase in strength was achieved, giving properties very similar to bullet proof steels.

A 1050°C heat treatment programme showed rapid sigma phase formation from δ ferrite but the process was reversed on further heating. This phenomenon was attributed to alloy migration as the weld microstructure approached equilibrium, the hardness of the alloys followed the variation in δ ferrite content being at their heightest at the beginning and end of heat treatment.

The bainite reaction in an Fe - 3.85 wt pct Si - 0.9 wt pct C steel is studied by several experimental techniques in the range of 250–450°C.

The high silicon content prevents the formation of cementite, so that the reaction is separated to two clearly distinct stages. In the primary stage ferrite forms alone, except at temperatures lower than 310°C where some carbides precipitate in it, and austenite becomes enriched in carbon. In the secondary stage occurring only above 400°C, the enriched austenite decomposes to ferrite and an unknown silicon carbide.

The microstructure, the enrichment of the austenite and the overall reaction kinetics of the two stages are studied and are found to be consistent with a displacive mechanism of the bainite reaction.

A tentative model, accounting for the competition of shear and diffusion, is proposed in order to fit our experimental data.