The atomic structure of dislocation cores in NiAl is studied by high-resolution transmission electron microscopy (HRTEM) and molecular dynamics (MD) calculations. Results are presented on dislocations with Burgers vectors b=a<100> and a<111>. A comparison with HRTEM image simulations indicates that the core of a 45° a <100> dislocation consists of Al atoms. The Burgers vector distribution shows a width of 2.2b. This corresponds very closely to MD results and is consistent with the relatively low Peierls stress of this dislocation. By detailed image analysis the angular dependence of the shear stress components of the dislocation are made visible. MD results obtained from 45° dislocations with opposite screw components suggest, that the helicity of the screw component might be discernible from high-resolution electron micrographs. A a<111> dislocation with <110> line direction is shown which exhibits a rather wide dissociation, probably into two a/2<111> partials.

Iron aluminides with the B2 structure are highly oxidation and corrosion resistant. They arethermodynamically compatible with a wide range of ceramics such as TiC, WC, TiB2, and ZrB2. Inaddition, liquid iron aluminides wet these ceramics very well. Therefore, FeAI/ceramic compositesmay be produced by techniques such as liquid phase sintering of powder mixtures, or pressureless meltinfiltration of ceramic powders with liquid FeAl. These techniques, the resulting microstructures, andtheir advantages as well as limitations are described. Iron aluminide composites can be very strong.Room temperature flexure strengths as high as 1.8 GPa have been observed for FeAl/WC. Substantialgains in strength at elevated temperatures (1073 K) have also been demonstrated. Above 40 vol.% WCthe room temperature flexure strength becomes flaw-limited. This is thought to be due to processingflaws and limited interfacial strength. The fracture toughness of FeAl/WC is unexpectedly high andfollows a rule of mixtures. Interestingly, sufficiently thin (<1 μm) FeAl ligaments between adjacentWC particles fracture not by cleavage, but in a ductile manner. For these thin ligaments the dislocationpile-ups formed during deformation are not long enough to nucleate cleavage fracture, and theirfracture mode is therefore ductile. For several reasons, this brittle-to-ductile size transition does notimprove the fracture toughness of the composites significantly. However, since no cleavage cracks arenucleated in sufficiently thin FeAl ligaments, slow crack growth due to ambient water vapor does notoccur. Therefore, as compared to monolithic iron aluminides, environmental embrittlement is dramaticallyreduced in iron aluminide composites.

The infrared joining of titanium-aluminides Ti50A150, Ti60Al40 and Ti70A130 at Tw= 1100∼1200°C for 30–60sec using Ti-15Cu-15Ni foil as brazing filler-metal was investigated. Multilayered structures are formed by isothermal solidification following solid-state interdiffusion. The diffusion of Al atoms is the main controlling factor pertaining to the microstructural evolution of the joint interface. Seven characteristic zones at Tw can be distinguished in the Ti50Al50 joint: γ-TiAl (I and II), α + β two-phase mixed, high Al% α-phase, α2-Ti3Al, β-Ti and residual liquid filler-metal. Five characteristic zones at Tw, are obtained in the Ti70A130 joint: α2-Ti3Al, α2 + β two-phase mixed, α + β two-phase mixed, β-Ti and residual liquid filler-metal. The observed joint microsturctures at room temperature are obtained from the phase transformation of these well-established high-temperature phases during cooling. A step-by-step microstructural evolution mechanism at Tw = 1150°C is proposed individually for Ti50A150 and Ti70A130 alloys. These steps are in good agreement with the observed microstructures and are consistent with the multiphase diffusion theories in solid-state systems. The microstructural evolution of Ti60A140 joint interfaces can also be explained by the proposed step mechanisms for Ti50A150 and Ti70A130 alloys.

A simple empirical model for the ideal cleavage energy, resulting from a rigid-body separation, is proposed in terms of four variables, viz., the elastic stiffness constant, the interplanar spacing, and two adjustable length parameters. The ratio of these parameters is assessed based on the available results of ab-initio slab-supercell calculations. Ideal cleavage energies and stress intensity factors of transition-metal silicides are estimated, and the available fracture toughness data are discussed.

Atom probe microscopy has been used to investigate elemental partitioning and segregation behavior in a TiAl-based alloy with a variety of alloying additions including Cr, Nb, W and B. These results indicate that in a stress-relieved state (2 h at 900°C) and a reheated state (2 h at 900°C, 2184 h at 800°C and 2 h at 1210°C) chromium, and to a lesser extent tungsten, is partitioned to the α2 phase. However, in an annealed state (2 h at 900°C and 720 h at 800°C), these elements are partitioned to the, γ phase. Segregation of chromium and tungsten to lamellar interfaces is observed in the stress-relieved material, but significant segregation was not observed in material subjected to the other heat treatments. A W- and B-enriched precipitate was observed in the reheated material and provides a possible explanation for the low tungsten concentrations measured in the matrix phases.

Dislocation dynamics in Ni3AI intermetallic single crystals has been studied by mechanical spectroscopy between 300 and 700 K. It has been found that in the anomaly domain of the flow stress, which is characteristic of this material, the mechanical loss of predeformed specimens is strongly dependent on strain amplitude, predeformation level, annealing temperature and time. The results can be interpreted as a combination of two phenomena which simultaneously occur as temperature is increased from 300 K to about 500 K: exhaustion of the mobile dislocation segments (superkinks) and pinning of the screw dislocation segments via cross-slip from the (111) onto the (010) planes.

A two-phase, NbCrTi alloy (bce + C15 Laves phase) has been developed using several alloy design methodologies. In efforts to understand processing-microstructure-property relationships, different processing routes were employed. The resulting microstructures and mechanical properties are discussed and compared. Plasma arc melted (PAM) samples served to establish baseline, as-cast properties. In addition, a novel processing technique, involving decomposition of a supersaturated and metastable precursor phase during hot isostatic pressing (HIP), was used to produce a refined, equilibrium two-phase microstructure. Quasi-static compression tests as a function of temperature were performed on both alloy types. Different deformation mechanisms were encountered based upon temperature and microstructure.

The characterisation of dislocation mechanisms in connection with macroscopic mechanical properties are usually performed through transient tests, such as strain-rate jumps, load relaxations or creep experiments. The present paper includes a careful and complete theoretical analysis of the relaxation and the creep kinetics. We experimentally show that the plastic strain-rate is continuous at the transition between constant strain-rate conditions and both load relaxation and creep test. The product of the plastic strain-rate at the onset of the transient test () with the characteristic time (tk) of the transient is found to be independent of , as theoretically expected. This is a clear indication that the assumptions underlying the theoretical analysis are relevant.

Composites based on Nb-Si are attractive candidates for use as structural materials at the very high temperatures required for future aircraft engines. The composites described were produced by directional solidification, which gives a microstructure consisting of Nb dendrites with an Nb3Si-Nb eutectic. The aim of this paper is to provide a detailed characterization of precipitates observed in the Nb dendrites in both binary and higher-order alloys. The precipitates possess the Nb3Si stoichiometry, but not the stable Nb3Si structure. The precipitates form a metastable orthorhombic crystal structure which is related to the Nb matrix via a simple orientation relationship.

Thermal defects in B2 FeAl samples with compositions between 49.5 and 54.7 at.% Fe were investigated using perturbed angular correlation of gamma rays (PAC). Vacancies on the Fe-sublattice were detected through quadrupole interactions induced at adjacent 111In/Cd probe atoms on the Al-sublattice. Five high-frequency quadrupole-interaction signals were detected (greater than 50 Mrad/s) that are attributed to complexes involving 1, 2, 3, 4 and (with less certainty) 5 Fe-vacancies in the first neighbor shells of the probes. These attributions are based on (1) a comparison between measured quadrupole interaction parameters and point-charge calculations of electric-field gradients for possible vacancy-probe complexes; and (2) numerical simulation of the evolution of site fractions of probes in the complexes at lower temperatures. Measurements were made at temperatures up to 950 C. Assuming that the equilibrium high-temperature is the triple defect (2 Fe-vacancies and one Fe-antisite atom), measurements over the range 600–900 °C yield a formation enthalpy of 1.1(1) eV for the triple defect. Below about 600 °C, Fe-vacancies are quenched-in with a fractional concentration of the order of 1 at.% close to stoichiometry. However, quenched-in vacancies continue to jump over short distances and trap next to the impurity probes atoms in day-long measurements down to 200 °C. Simulations of site fractions below 700 °C were used to determine binding enthalpies of vacancies with probe complexes. Binding enthalpies obtained for the first four vacancies were 0.23, 0.23, 0.15 and 0.18 eV. Simulations in the range 200–700 °C suggest a negative value for the formation entropy. The negative value indicates either that triple defects stiffen the B2 lattice or that repulsive defectdefect interactions become important at the high defect concentrations in FeAl.

Two methods of repeated transients (stress relaxation and creep) which can be performed during straining at a constant strain rate are recalled together with the way the results are interpreted. The repeated creep method has been adapted to Ni3A1 compounds for which the amount of creep strain during the second creep could not be detected so far. Because of a small stress increase between creeps, it is possible to measure microscopic activation volumes, dislocation exhaustion rates and strain hardening coefficients. Results in Ni3A1 polycrystals and Ni3(AI,Hf) single crystals are presented. Orders of magnitude of dislocation exhaustion rates are given. They are particularly high in Ni3A1 as compared with other materials. They compare well with the work hardening coefficient and with total dislocation densities measured by electron microscopy.

Brittle-to-ductile transition (BDT) temperature (TBD) was evaluated according to temperature dependence of tensile properties under different strain rates from 10−5to 10−1s−1in two-phase Ti-47Al-2Mn-2Nb and Ti-47A1–2Mn-2Nb-0.8 TiB2alloys with nearly lamellar microstructure. Based on the strain rate dependence of the determined TBD values, apparent BDT activation energies were determined using Zener-Hollomon factor. Tensile fracture surfaces were observed using a scanning electron microscope while deformation substructures were investigated by transmission electron microscopy. It was found that the BDTT of both alloys increased sharply with the strain rate and that the minor addition of 0.8 vol% TiB2reduced TBD by about 100K at the same strain rate. The TiB2addition also decreases the apparent BDT activation energy from 324 to 256 kJ/mol. Both of these values approximate to self- or inter-diffusion of Ti and Al atoms in TiAl phase. Transgranular fracture and dimple fracture were found dominant in fracture surfaces below and above TBD, respectively. The most common 1/2<110] ordinary dislocations were found to begin climb at mound TBD. All this evidence, as well as a theoretical calculation using the Nabarro Model, add up to a conclusion that the BDT is controlled by dislocation climb in both alloys.

It is expected that the κ-phase of the intermetallic compound Co3A1C0.5 would strengthen Cobase alloys used at high temperatures like the γ' -phase of Ni-base superalloys. Tensile and creep rupture properties of Co+κ two-phase alloys with κ-phase volume fractions up to 0.75 were investigated. Alloy samples made by directional solidification casting were annealed at 1573 K for 3.6 ks and at 1373 K for 28.8 ks in vacuum, followed by Ar gas cooling. Tensile tests at RT and 1073 K and creep rupture tests at 1089 K under a stress of 172 MPa were conducted with the tensile axis parallel to the solidification direction. In alloys with low κ-phase volume fraction, cuboidal K-precipitates with average particle diameters of 0.4 to 1.0 μm were observed. They were coherent with the Co(fcc) matrix with misfits of about 3%. As the κ-phase volume fraction increased, tensile strength also increased. The alloy with κ-phase volume fraction of 0.4 had a 0.2% proof stress of 817MPa, tensile strength of 1047 MPa at RT, creep rupture life of 1.43 Ms, and tensile strain higher than 10%. These strengths are better than those of the conventional Co-base alloys. However, ductility of alloys with κ-phase volume fraction larger than 0.4 decreased due to large eutectic and primary κ-phase particles.

Vacancy-like defects in NiAl in the composition range 47 at.-% < CNi < 53 at.-% are investigated by means of positron lifetime spectroscopy and Doppler-broadening measurements. The observed lifetimes in the annealed samples confirm that defects are quenched-in during the production of the samples. Isochronal annealing of samples quenched at 1600°C and after proton irradiation show that the induced defects are quite different.

Mechanical alloying and plasma assisted sintering are used to produce nanocrystalline Ti-Al alloys. Microstructural characterization and measurement of mechanical properties are reported. The alloys investigated include AITi+X and Al3Ti+X where X represents either Cr, Mn or Fe. Mechanical alloyed powders have a microstructure, which consist of an amorphous matrix containing small crystalline domains. Sintered specimens have a crystalline microstructure with nano-sized grains. Alloys based on AlTi-X are composed mainly of the phases α 2 (DO19) and γ (Ll0) distributed homogeneously in the form of nanograins (100–300 nm) resembling a globular structure. Al3Ti-X alloys show a cubic ordered phase (L12) with grain sizes ranging from 16 to 40 nm. Vickers hardness measurements of sintered alloys show a considerable increase with respect to as-cast alloys. Compression tests show a high mechanical strength of these alloys, although in the case of A13Ti-X, no ductility is found.

The work hardening behaviour of gamma base titanium aluminides was investigated by mechanical testing, electron microscope observations and recovery experiments. The main objectives of the paper are: (i) to ascertain the nature of work hardening at room temperature, (ii) to identify deformation induced glide obstacles which can be overcome with the aid of thermal activation, (iii) to assess the thermal stability of deformation induced defect structures.

We studied the high temperature corrosion of spray atomized and deposited FeAl40at% based intermetallic alloys immersed in a molten salt mixture of 80%V2O 5+20%Na2SO4 (wt%) over the temperature range of 600–900°C. Experiments were realized by the weight loss method and the potentiodynamic polarization electrochemical technique in three different samples: FeA140at%, FeA140+0.lat%B and FeA140+0.lat%B+10at%A12O3. Measurements of weight loss and corrosion current density as a function of the molten salts temperature were obtained and discussed in terms of the passive layer morphology and corrosion products formed during the tests. It was found that the iron aluminide doped with boron and reinforced with alumina particulate was more corrosion resistant in the test temperature range. The weight loss experiments revealed that at 700°C all alloys developed maximum corrosion rate. This behavior was related with the dissolution of protective oxide layer on metal base due the formation of vanadate phases which are highly corrosive at this temperature.

For the improvement of mechanical properties, especially ductility, of intermetallic compoundbased alloys, microstructural control has been adopted to binary and higher-order systems. Although lamellar structure formed by a eutectic reaction has been attracting broad attention to ductilize alloys with a combination of brittle/ductile or brittle/brittle phases, it should be pointed out that the microstructure such as lamellar size or colony size is hard to be controlled after the solidification process. In that viewpoint, high-temperature solid-solid phase transformation provide significant advantages in handling and large super-cooling for the control of lamellar structure through eutectoid reaction.

The Ni3Ge-Fe3Ge model system provides us with a unique opportunity to characterize the mechanisms of deformation in both anomalous and normal L12 intermetallic alloys. The elastic moduli of alloys in this system have been measured and used as benchmarks for first principles calculations. At 77K, increasing the Fe content has been found to result in a dramatic increase in flow stress. The Ni-rich alloys exhibit a yield strength anomaly, but as Ni is replaced by Fe, the anomalous temperature dependence gradually disappears, and no yield strength anomaly is observed for alloys with more than 25 at% Fe. At low temperatures and Fe contents, the deformation microstructure has been found to be dominated by Kear-Wilsdorf locking; but a transition from octahedral glide and Kear-Wilsdorf locking to cube glide is observed as either Fe content or temperature is increased. This transition is related to changes that occur in the core structures of dissociated superdislocations and planar fault energies measured through computer simulations of weak-beam TEM images.

The low ductilities of FeAl alloys led us to explore powder metallurgical processing technology to obtain sheets of 0.2mm thickness as opposed to manufacturing processes based on hot rolling of cast FeAl alloys. In our approach, water atomized FeAl powders were roll compacted to 0.66mm with a polymeric binder using two counter rotating rolls to a green density of 3.1 g/cc. Roll compacted green sheets were then de-bindered in nitrogen in the temperature range of 300 to 600°C for several hours prior to sintering the sheets in vacuum. Sintered sheets were rolled down from 0.66 to 0.20 mm in three different stages resulting in a total reduction of 69% Vacuum annealing of the sheets was carried out between each stage of the reduction process to eliminate edge cracking associated with the work hardening of the FeAl. The properties of the FeAl sheets depend on the Al content, annealing temperature and time in a vacuum furnace. The fine microstructure of FeAl sheets led to tensile elongations of 4 to 6%. The sheets are formable at room temperature, and possess excellent mechanical properties both at room and high temperatures.