Ordered Intermetallics — Physical Metallurgy and Mechanical Behaviour

Current sophisticated electronic structure simulations are at the forefront of understanding and predicting a variety of materials properties of intermetallic compounds. Several examples are given here that illustrate how first principles total energy local density methods have addressed the problems of (i) phase stability and the effect of ternary additions, (ii) anti-phase boundaries (APB’s) in B2 NiAl, FeAl and RuAl aluminides and other faults in determining their structural and bonding character. A key objective has been to attempt to understand, at the electronic level, fundamental quantities that may be related to the crucial ductility issue in high temperature intermetallics. Differences between observed ductility properties of related systems may relate to their differing electronic and bonding properties, particularly the nature of p-d hybridization and the directional charge distributions of the states near the Fermi energy.

We review, briefly, the first principles KKR-CPA theory of the electronic structure and energetics of alloys in which compositional disorder plays a role. We also review the first principles theories of ordering and alloy phase stability that are built on the KKR-CPA description of the disordered state. We point to a number of underlying electronic driving mechanisms of ordering and clustering that have been uncovered. Specifically, we emphasize the important role that Fermi surface nesting plays in driving specific instabilities. Using a newly developed method for treating the effects of disorder in alloys that have complex lattices we present results for the ordering energies of beta phase Ni_cAl_(1-c) alloys. Finally, we show alloy “Fermi surfaces” in β-phase Ni_0.625Al_0.375 that support the notion that the pre-martensitic phenomena observed in this alloy are Fermi surface driven.

A brief non-technical overview is presented for first-principles calculations of alloy phase equilibria. Merits and drawbacks of various methods are briefly discussed.

The search for new pseudobinary intermetallics with a required structure type and mechanical properties can be guided at the outset by the use of phenomenological two-dimensional structure maps which order the known structural data base on binary compounds. This paper reviews the progress made and problems encountered in using the maps in developing new cubic transition metal aluminides. In particular, it emphasizes the need for a microscopic quantum mechanical understanding of why, for example, Al₃Nb cannot be stabilized in the cubic form whereas Al₃Ti can, and why the latter remains brittle even though single crystals of Ni₃Al with the same cubic L1₂ crystal structure are ductile.

A cluster expansion is used to compute fcc ground states from first principles for the Pd-V system. Intermetallic structures are not assumed but derived rigorously by minimizing the configurational energy subject to linear constraints. A large number of concentration-independent interactions are calculated by the method of direct configurational averaging. Agreement with the fcc-based portion of the experimentally-determined Pd-V phase diagram is quite satisfactory.

Structural and thermomechanical properties which include lattice constants, bulk moduli, Debye temperatures, and Grüneisen constants for the elements and ordered compounds can be obtained accurately using total energy electronic structure calculations within the local density approximation. Examples of these calculations are given using the linearized muffin tin orbital method (LMTO) for the Nb-Ru and Ru-Zr binary systems. Volume dependent pair- and many-body chemical interactions are extracted from the total energy results. These are used with the cluster variation method (CVM) to calculate the solid state equilibrium phase diagram by incorporating local volume relaxation and vibrational free energy using the Debye-Grüneisen approximation.

The computer calculation of phase diagrams has been developed in the last two decades starting with the pioneering work of Kaufman et al.. His approach was applied and extended by a number of research groups known under the logo of CALPHAD. Most of their work can be found in the CALPHAD journal and at the occasion of their annual CALPHAD meetings. The purpose of these activities was and still is to provide a tool for estimating the phase equilibria in multicomponent systems. Usually the amount of experimental data is meagre and comes from different areas, e.g. phase diagram data, calorimetric and general thermophysical measurements, structural data. The computer programs are constructed in such a way that they can bundle all these different kinds of informations in an optimizer module to arrive at numerical values for adjustable parameters which define the thermodynamic functions. These functions have been developed in the past and this activity goes on. We may distinguish three types of approaches: (I) the pure formal mathematical approach consisting in a polynomial representation of the thermodynamic functions (CALPHAD approach), (II) the modeling of thermodynamic functions, (III) statistical thermodynamics treatments.The presented paper tried to give an overview of these techniques with emphasis on the special aspects of materials development based on intermetallic phases. Special aspects were: How good is the description of intermetallic phases? What about the homogeneity range of intermetallic phases? What about the predictive power of the techniques, e.g. what can be predicted in a ternary system from existing knowledge in the binary subsystems? These aspects were discussed and illustrated with particular examples out of the huge amount of existing data in the literature as well as unpublished new results.The interested reader is refered to the references listed below. This list cannot be complete. A selection has been made such that the main strategies can be found and that typical aspects of phase diagrams of intermetallic systems are treated.

The status of atomic ordering in ternary and quaternary III–V compound semiconductors is reviewed briefly. Evidence favors the occurrence of CuPt-type ordering in these materials. Reconstruction at (001) surfaces is essential for the formation of the ordered structure that is metastable in the bulk. Ordering reduces bandgaps, and degradation resistance of light emitting devices is enhanced.

The multitude of critical phenomena which occur at surfaces and interfaces is briefly reviewed from a theoretical point of view. Three types of critical effects are distinguished related (i) to the 2-dimensional character of the interface, (ii) to its morphology, and (iii) to its structural changes at phase transitions in the bulk. Wetting phenomena in three dimensions belong to the last category (iii). Some recent theoretical results for such phenomena are also discussed: (a) Wetting of a moving interface; (b) Wetting in the 3-dimensional Ising model; and (c) Wetting of an inhomogeneous substrate.

Antiphase Domain Boundaries play an important role in phase changes and microstructural stability of ordered alloys and intermetallics as well as affecting mechanical behaviour. The origin of APB faults in annealed material is examined here, and in particular the differences between a sharp boundary, as produced by crystal shear, and a relaxed fault structure are emphasized. The kinetics of relaxation of a shear-produced fault are examined and it is shown that fast relaxation may greatly affect the movement of dislocations by creating locking stresses as well as affecting cross slip behaviour, and hence significantly affecting mechanical properties.

Intermetallic compounds having the L1₂ structure show many unusual mechanical properties. For example, many show an anomalous temperature dependence of the flow stress which has been related in various models to the structure of screw dislocation cores. Others, however, show a bcc-like behavior, which also has been related to core structures. Apparently unrelated to the flow behavior, L1₂ compounds also have a tendency for brittleness at low temperatures, with failure either occurring by transgranular cleavage or by intergranular failure mechanisms. Cleavage in some L1₂ materials, e.g. Fe-modified Al₃Ti, tends to occur on a remarkably large number of planes in both poly- and single crystals. The cleavage resistance of this material can be slightly increased only by Mn additions. Ni₃Al, on the other hand, tends to fail intergranularly, unless the grain boundaries contain B, and even then, failure under triaxial stress states tends to occur intergranularly. Intergranular cracks propagate randomly through the grain boundaries of Ni₃Al with little preference for boundaries of any particular type, except that cracks tend to avoid symmetrical twin and low angle tilt boundaries. In this paper our current understanding of the flow and fracture behavior of this class of materials will be reviewed.

The fundamental aspects of deformation and fracture behavior of ordered intermetallics were investigated on the basis of quantum mechanical total-energy calculations and anisotropic elasticity theory for dislocations and cracks. These first-principles calculations were based on the local-density functional (LDF) theory. The LDF equations are solved either by full-potential linearized augmented plane-wave (FLAPW) method or by mixed-basis pseudopotential method. Our approach represents a major advance in applying LDF to solids, in which the LDF equations are solved without any shape approximation to the potential or charge density and a high degree of variational freedom (and precision) can be achieved. The calculated elastic constants, various shear fault energies, defect self-energies, and cleavage energies were used in conjunction with the continuum modeling of dislocations and cracks to predict the mechanical behavior and to understand the underlying electronic mechanism of observed mechanical properties.

Surface defects determine the extension of dislocation cores. Information on APBs and stacking faults is now available in a large variety of structures and compositions of intermetallics alloys. It is increasingly evident that real APBs are liable to deviate from geometrical ones.

The structures and properties of dislocations and other defects in deformed L1₂ ordered Ni₃Al and Ni₃Ga are observed with transmission electron microscopy. The principal intention of this paper is to examine the changes in the structure and behaviour of dislocations in a wide temperature range (-196 to 900 C) with a view to understanding the transitions of slip systems and yielding properties. Dislocations with low mobilities in the various temperature regions are first identified with weak-beam microscopy and the core structure of these dislocations are then observed directly by lattice resolution TEM. In order of increasing temperature, dislocations observed to have non-planar structures are: 30° >112</3 super-Shockley partials (Giamei locks), screw >101< superdislocations (Kear-Wilsdorf locks), edge >101<(001) superdislocations (Lomer-Cottrell type locks), and 45° >100<(001) dislocations (B5 locks). The effects of temperature on the formation and destruction of these locks are discussed and related to the transition of slip systems and the change of yield properties. The operation of a kink mechanism for the unlocking of Kear-Wilsdorf locks is shown to lead to the formation of special kink configurations with switched partials and the formation of antiphase domain boundary tubes which are also observed to characterize the region of the yield stress anomaly.

A new theory is presented to explain the yield stress anomaly for (111) slip in L1₂ alloys. It is shown that strong sessile dipole barriers are formed by the superkinks at the ends of screws which have cross-slipped from (111) to (010). These barriers stabilise the cross-slipped screws, which slip further to form Kear-Wilsdorf locks. The yield stress is controlled by the unlocking of the superkinks, which bypass the screws and Kear-Wilsdorf locks, and which generate new mobile screws which become locked again. The unlocking mechanism is thermally activated with a large athermal component. The theory accounts for the mechanical properties, including the small strain-rate sensitivity of the yield stress, and explains many of the electron microscope observations. The application of the theory to deformation of thin foils in the electron microscope is also discussed.

In the last few years, substantial progress has been made in our understanding of the microstructure and the deformation and fracture behavior of TiAl compounds. We have made uniaxial tension tests of specimens of the TiAl compounds whose lamellar orientation is controlled so that shear deformation occurs parallel to the lamellar boundaries only in the TiAl lamellae. Then, the TiAl phase itself, which is in equilibrium with the Ti₃Al phase, has been found to be easily deformable, and yet brittle fracture occurs in the TiAl phase when the specimens fail. This paper reviews such a deformation and fracture behavior of the TiAl phase together with some key aspects of the progress that was made on TiAl compounds with the TiAl/Ti₃Al lamellar microstructure.

The brief development history of intermetallic alloys as structural materials has led to several examples which illustrate the important effects of alloy chemistry on the flow and fracture properties of these alloys. Several recent investigations have begun to investigate the theoretical basis underlying such phenomena. However, such treatments have typically not addressed the aspects of these property variations which may be driven by solute-dislocation interactions. Within dilute substitutional solid solutions of intermetallic alloys, several distinct glide systems can often be identified over a range of deformation temperatures. For each of these the distinct solute-dislocation interactions must be quantitatively described, related to the mobility of dislocation loop segments, and then combined in a general way to describe the observed alloy behavior. Within the present manuscript, several examples of chemically-dependent deformation phenomena are reviewed. Following this, solute hardening effects in Ni₃Al alloys are reviewed to illustrate specific mechanisms which must be investigated and described by solute-strengthening theory. Finally, some results from anisotropic elasticity calculations and atomistic simulations within the embedded atom formalism, of solute-dislocation interactions in Ni₃Al alloys, are presented and discussed.

Recent studies of fatigue behavior of intermetallic alloys under both stress and strain control conditions are reviewed. Effects of composition, temperature, frequency and environment on high cycle fatigue, crack growth and low cycle fatigue are discussed. In addition, crack initiation phenomena in Ni₃Al single crystals are described and a new model of fatigue crack initiation is outlined.

γ-titanium aluminides exhibit a deformation behavior like normal fcc metals from 4K to RT, but an anomalous strain hardening from RT up to 1073 K similar to L1₂ Ni₃Al. Unlike Ni₃Al, γ-titanium aluminides have two different slip systems: 1/2[110] (111) and [101] (111) slip, both of which are responsible for a strong hardening behavior at high temperatures. Although little has been known about the mechanisms causing the hardening behavior, preliminary results indicate that the non-planar structures of the dislocations in the TiAl appear to be a key to the hardening at the low temperature regime (lower than 800 K) while cross-slip of the APBs of the superdislocations onto cube planes and the cross slip or climb of the ordinary dislocations onto non-glide planes are responsible for the hardening above 800K. Other potential mechanisms for the hardening are discussed. The composition dependence of TiAl on mechanical properties at RT in various ternary γ-titanium aluminides is reviewed.

Models on cleavage fracture, interfacial fracture, and on the brittle-to-ductile transition are overviewed for intermetallics and compared with recent experimental results on the rate and temperature dependences of the fracture toughness of intermetallics. The influence of grain size, phase distribution, temperature and environment on the fracture toughness of NiAl and TiAl based alloys was measured and discussed. It was found that the brittle-to-ductile transition temperature in single phase NiAl alloys cannot be lowered below the transition temperature of suitably oriented single crystals by grain refinement. But the toughness of two-phase alloys is improved when the ductile phase completely surrounds the brittle phase in sufficient thickness. The fracture toughness of many intermetallic alloys was found to be extremely rate sensitive. For NiAl (B2) single crystals the influence of orientation, strain rate and temperature was measured and discussed in view of recently developed dynamic models of the brittle-to-ductile transition.

Recent studies have demonstrated that moisture-induced environmental embrittlement is a major cause of low ductility and brittle fracture in ordered intermetallics with high crystal symmetries (e.g., L1₂ and B2). The embrittlement involves the reaction of reactive elements in intermetallics with moisture in air and the generation of atomic hydrogen at crack tips. The loss in ductility at ambient temperatures is generally accompanied by a change in fracture mode from ductile appearance to brittle grain-boundary fracture in many L1₂ intermetallics, and to brittle cleavage in body-centered cubic (bcc)-ordered intermetallics. In a number of cases, the embrittlement was alleviated by alloy design through control of microstructure and alloy composition.

Intergranular fracture in L1₂ compounds such as Ni₃Al poses two basic questions. First, whether the grain boundary brittleness is intrinsic and if yes what is its origin and why it is not found L1₂ alloys such as Cu₃Au. Second, how and why is the brittleness affected by alloying and deviations from stoichiometry. These problems are discussed here in the light of the results of the Monte Carlo atomistic studies of the structure and composition of grain boundaries. For stoichiometric Ni₃Al virtually no compositional disorder occurs even at very high temperatures, while in Cu₃Au a significant disordering takes place already at room temperature. As suggested earlier [27, 30, 31], preservation of the compositional order in grain boundaries may be the principal reason for their intrinsic brittleness and, therefore, these results may explain why grain boundaries are brittle in Ni₃Al but not in Cu₃Au. In non-stoichiometric Ni₃Al the compositions of grain boundaries are very different in nickel rich and aluminum rich alloys, respectively. Whereas both components segregate to the boundaries when in surplus, nickel invokes disordering while aluminum does not. This may be the reason why the intergranular brittleness can be alleviated by additional alloying, for example by boron, in nickel rich alloys only.

The local compositional order and dislocation structure of grain boundaries in Ni₃Al, with and without boron, were examined using electron microscopy techniques. Lattice imaging studies showed that small angle twist, tilt and mixed boundaries and large angle (near Σ = 5) [001] twist boundaries are ordered up to very close to the interface plane. A compositionally disordered region ~1.5 nm thick is present in the vicinity of a large angle general boundary in boron-doped Ni₃Al, while similar boundaries in boron-free material are ordered. Image simulations were performed and it was shown that the experimental observations of a locally disordered region cannot be explained as being an artifact. Dislocations with Burgers vectors that correspond to anti-phase boundary (APB)-coupled superpartials were found in small angle [001] twist boundaries in both boron-free and boron-doped Ni₃Al and a small angle [011] tilt boundary in boron-doped Ni₃Al. The APB energies determined from the dissociation of the boundary dislocations were smaller than reported for bulk Ni₃Al. For small angle twist boundaries the presence of boron reduced the APB energy at the interface until it approached zero.

It is argued that the improvement in the ductility of Ni₃Al, Ni₃Ga, Ni₃Si and Ni₃Ge through the addition of boron (in the ppm range) and of Zr₃Al through fast neutron irradiation is related to an increase in the transmission of slip across the grain boundaries. The transmittal mechanism is dislocation nucleation at the heads of pile-ups, and this is made easier through the disordering of the boundaries. The evidence supporting this view is reviewed. It is suggested that grain boundary disorder may be a general requirement for ductility.

This article involves a number of studies on the alloying effects on intergranular fracture of L1₂ ordered intermetallics, and demonstrates that their ductility strongly depends upon atomistic composition and associated crystal and electronical structures at grain boundaries. Component atoms with large size, which occupy anti-structure site and disordered sites, and their preferential sites as third (or quaternary atoms), are shown to influence the grain boundary strength and fracture behavior in moderate levels of additions. Addition of interstitial atoms such as boron, carbon and beryllium with small size is extremely effective to enhance or to reduce the grain boundary cohesive strength in trace amount of levels in matrix but in highly enriched levels at grain boundaries. Also, gaseous hydrogen atoms with small size, which is mobile and penetrated from environment at room temperature, are shown to dynamically reduce the grain boundary strength, resulting in the environmental embrittlement. It is suggested that the grain boundary strength and therefore ductility of L1₂ ordered intermetallics can be controlled by compositions of components and interstitials, state of ordering and prohibition of penetration of hydrogen.

Essential features of the microstructures which can be produced via phase transformations in two-phase alloys based on γ-TiAl and α₂-Ti₃Al are described. Effects of microstructure on the deformation and fracture behaviour are discussed with a view to finding possible pathways towards improved ductility and toughness in such alloys.

Theoretical aspects of diffusion in intermetallic compounds are discussed. Order-disorder compounds as well as compounds highly ordered at any temperature are considered. Theory is compared to the experiment.

The diffusion coefficients of A and B atoms in the B2 type AB alloy are derived for the case that highly correlated vacancy jump cycle (six-jump vacancy cycle) is operative. The effective jump frequency to complete the cycle is expressed in terms of frequencies for individual vacancy jumps; the calculation is made by applying the concept of the mean first passage time known in the theory of stochastic processes. On the basis of the analytical expression for the diffusion coefficients, the ratio of the diffusivities of the two constituent atoms and the isotope effect can be discussed quantitatively.The correlation effect in the six-jump cycle can be regarded to consist of two effects: a microscopic and a macroscopic effects. The former refers to the correlation between individual jumps involved in the cycle, for which the correlation factor can be neither defined nor calculated.

The phenomenological basis for the description of interdiffusion in multicomponent systems and the experimental approaches employed for the determination of interdiffusion fluxes, interdiffusion coefficients and zero-flux planes are briefly reviewed. Interdiffusion coefficients determined for β (bcc) Fe-Ni-Al alloys at 1000°C are presented and the development of zero-flux planes is illustrated with selected Fe-Ni-Al diffusion couples. A new analysis involving the concepts of average, effective interdiffusion coefficients and penetration depths for the individual components in a diffusion couple has been developed. Diffusion paths for a few interdiffusion experiments carried out in the Ti-Al-Nb system at 1100°C are also presented. A Ti-Al-Nb diffusion couple that developed B2 phase as a diffusion layer is analyzed for the determination of effective interdiffusion coefficients for the components over selected concentration ranges; these data are discussed in the light of ordering in the B2 phase.

The experimental data for A-diffusion (A=Ni) and for B-diffusion (B=Sb, Sn, In) in A/B-alloys with the B8₂-structure have been compared with a number of possible diffusion mechanisms. The diffusion behaviour is well understood in terms of a main diffusion mechanism, using the double tetrahedral interstices in the case of A-diffusion, and a pure vacancy mechanism in the B-sublattice for B-diffusion. Minority mechanisms explain the temperature variation of the D_┴/D_‖-ratio.

Quasielastic neutron scattering (QNS) and quasielastic Mossbauer spectroscopy (QMS) permit to deduce the jump vector of diffusing atoms. This is possible by comparing the angular dependence of quasielastic line broadening with model predictions. Results are reported for Ni diffusion in NiSb (B8) and Ni₃Sb (DO₃) and for Fe diffusion in FeAl (B2). For NiSb we conclude that the Ni atoms jump alternately between regular and interstitial sites. Ni₃Sb and FeAl contain high concentrations of vacancies; conclusions on the possibilities for jumps via the vacancies are drawn.

In the first part, the ordering kinetics of reversibly ordered phases such as Ni₄Mo and Ni₃Fe are examined; these can be initially disordered either by quenching from above T_c (which < T_m for a ‘reversibly’ ordered alloy) or by cold-working the ordered alloy. The relationship of ordering kinetics to diffusivities is outlined. The interrelation of ordering kinetics and the kinetics of recovery and recrystallization of reversibly ordered alloys is briefly discussed. — In the second part, the reordering of Ni₃Al, an alloy of the ‘permanently’ ordered variety, is analysed. This alloy is ordered, in equilibrium, up to the melting temperature; it can be substantially disordered either by severe cold work or by ultrarapid quenching. Some recent work on disordering of Ni₃Al by ball-milling is described; the process is associated with unexpected changes in lattice parameter. The effect of annealing and gradually reordering such a mechanically disordered Ni₃Al upon its hardness is briefly treated.

In the first section of this overview the creep behaviour of single-phase intermetallic alloys is discussed with respect to stress dependence, temperature dependence and effects of composition. The second section refers to two-phase intermetallic alloys, and both particulate and non-particulate alloys are regarded. Data are presented for single-phase and two-phase NiAl-base alloys, and the prospects for materials developments for application temperatures above those of the superalloys are briefly discussed.

The creep resistance and elevated temperature deformation mechanisms in CoAl, FeAl, and NiAl are reviewed. The stress and temperature dependencies of the steady state creep rate, the primary creep behavior, the dislocation substructure, and the response during transient tests are used as the main indicators of the deformation processes. In single phase intermetallics, the influence of grain size, stoichiometry, and solid solution hardening have been examined. In addition, the effect of adding dispersoids, precipitates, and other types of reinforcements to improve creep strength are compared.

Several intermetallic compounds with the L1₂-structure exhibit a yield strength anomaly (YSA), i.e., the yield stress increases as the temperature increases. As shown for Ni₃Al, no corresponding “creep strength anomaly” exists. The creep strength decreases in a normal manner with increasing temperature. Its orientation dependence is different from that observed in the YSA regime. The transient creep of Ni₃Al is composed of a normal (“work hardening”) and an inverse (“work softening”) part. The reasons for such transients are discussed. Specific models for the dependence of strain rate vs. strain during inverse creep as well as for the orientation dependence of the creep strength are reviewed and compared with experiments. Experiments and models indicate that the strain rate ε depends on the strain ε approximately as ε or ε2/3. The orientation dependence of the creep strength is particularly pronounced at low temperatures and changes at high temperatures and low stresses. At low creep stresses, Ni₃Al is not as strong as high-strength superalloys, due to its low stress exponent. However, further optimization may be possible. As to the creep of other L1₂-compounds, very little information is available and no distinct evidence for inverse creep has been found. This may be due to the lower yield stresses and less pronounced YSA in these compounds.

This contribution is a review of the programs at the National Laboratories and various U.S. Universities that are engaged in research on intermetallic compounds and alloys that are either ductile or ductility can be induced by small additions of alloying elements. The coverage ranges from the growth of single crystals to the investigation of grain boundary effects on the atomic level.

The quest for advanced materials has renewed interest in documenting relationships between atomistic considerations and macroscopic properties. This effort in computational materials design spans the spectrum from angstroms to millimeters, from physics to materials science to mechanics, from first-principles to design criteria. Establishment of relevant relationships will often run into a gap of existing knowledge at some point of the spectrum. The pull across this gap must often be supplied by requirements from the more macroscopic community. One class of materials that has received much attention through this approach is the high temperature intermetallics. Ductility issues have been explored and some progress has been made. The gap of knowledge appears to be between the atomic scale properties that may be modeled and microstructural and micromechanical mechanisms that are influenced. Defining the gap helps to identify and prioritize key areas of research. The use of computational materials design for structural materials research and development is in its infancy. It is hoped that this discussion will encourage additional systems and problems to be identified and attacked by this approach.

In this article, research and development on intermetallics in Taiwan, R.O.C. is reviewed. Starting from 1984, R&D on intermetallics has been aiming at nickel aluminide, titanium aluminide and titanium nickel shape memory alloys. For Ni₃Al intermetallics, a series of effort has been focused on ductilization and strengthening by alloying and processing. Single crystal study and creep resistance for Ni₃Al intermetallics and its alloy modifications were emphasized in this topic. A mechanism of strengthening for long-range ordered L1₂ matrix by a dispersoid of disordered Ni₃(AlCr) is proposed. For the titanium aluminide systems, microstructure of α₂+γ two phase TiAl and texture-strengthening of Ti₃Al-Nb has been investigated. For titanium nickel SMA intermetallics, recent research interests on TiNi intermetallics are centered on the premartensitic transformation, transformation sequences, high temperature SMA and hydrogenation of NiTi alloys. Lastly, an innovative process for preparing any kind of intermetallics was developed and will be described briefly in this article.

The brief for this paper is to outline the intermetallics work in the UK, which is not covered in the presentations from Cambridge, Oxford and Imperial College at this meeting. The total research effort in the UK into intermetallics is not large and in this brief review an attempt is made to summarise other University-based work on defects in intermetallics, before describing some of the more industrially-based work. The industrially-supported work is aimed at producing alloys based on intermetallics and is centred on the new plasma melting facility, processing and testing facilities at the IRC. Finally the direction of future UK research into alloys based on intermetallics is briefly discussed.

This paper will provide an overview of the current status of the research and development of ordered alloys for high temperature structural applications, which were conducted in P.R.China. The physical and mechanical properties of ordered alloy will be reviewed with enphasis on aluminides. Current research programs in P.R.China will be outlined, and the expections of the future will be discussed.

In the present paper, potential and prospects of some intermetallic compounds for structural applications are discussed by emphasizing two specific points which seem to be of prime importance for the future research and development activities. The first point stresses upon the benefits associated with the development of multi-phase intermetallics and describes a novel approach for creating the two-phase γ-γ′ type microstructure in different alloy systems. The second point deals with the problem of stability of the constituent phases in these multi-phase materials. Views expressed in this paper are frequently corroborated by referring to our most recent results obtained in various alloy systems based on Ni₃Al, iron, niobium, Ti₃Al+X (X=Nb, V and Mo) and NiAl.

A survey of recent and current research on lightweight intermetallic titanium aluminides in Germany is presented. The main activities are centered around γ-TiAl based alloys and multi-phase alloys based on Ti₃Al and Ti₅Si₃ with Nb additions. The focus is on the interrelation between microstructure and mechanical properties.

NiAl alloys offer significant payoffs in gas turbine applications. Excellent progress has been made in understanding their mechanical behavior and improving low temperature ductility and high temperature strength. Significant improvements in mechanical properties have been obtained with microalloying. The next challenge is to develop an alloy which has the required balance of ductility, toughness, strength and other properties such as fatigue and impact resistance. Development of design, processing and test methodology for components made out of low ductility and anisotropic materials will also be required. While significant challenges remain, the prognosis for using NiAl alloys as high temperature structural materials is good.

The Advanced Research Workshop on Ordered Intermetallics—Physical Metallurgy and Mechanical Behavior provided a scientific forum to discuss and assess the research and development of ordered intermetallic alloys. The workshop consisted of four and a half days of feature presentations and discussions on recent advances in these areas. The last half day was devoted to assessment of current intermetallic research and recommendations of critical areas for future studies. The workshop focused on three specific areas: (1) electronic structure and phase stability, (2) deformation and fracture, and (3) high-temperature properties (including creep, diffusion, grain growth, etc.).