Low-pressure plasma deposition (LPPD) was used to synthesize an in-situ MoSi2/SiC composite using 100% methane (CH4) as a powder carrier gas. X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) identified MoSi2, Mo5Si3, Mo5Si3C, Si02, and SiC as the phases present in the composite. XRD and XPS revealed ∼6 vol% SiC in the as-sprayed material. Annealing of the as-sprayed composite increased the SiC content to ∼8 vol% while reducing the Si02 volume fraction. Transmission electron microscopy studies revealed a fine homogeneous distribution of SiC and/or carbide particles at prior splat boundaries in the MoSi2 matrix. Wavelength dispersive spectroscopy (WDS) confirmed the increased presence of carbon in the in-situ materials. Fracture toughness measurements yielded values on the order of 10 MPa m½ for annealed composites. The creep behavior of the LPPD reactive spray composite dramatically improved compared to unreinforced LPPD MoSi2. Additionally, the creep behavior was shown to be equal to or better than that of powder metallurgy MoSi2/SiC composites containing higher percentages of SiC.

Various powder mixtures were prepared by a modified mechanical alloying technique. Starting from elemental Mo-, Si- and C-powders the influence of milling conditions on phase formation during the milling process and the subsequent heat treatment was investigated. Phase formation during sintering and sintering kinetics of activated starting mixtures were studied by differential scanning calorimetry (DSC), thermal graphimetry (TG), X-ray diffraction (XRD) and dilatometry. The results show that phase formation during milling or sintering strongly depends on milling conditions. Optimized powder mixtures of single phase and reinforced molybdenum silicides show high densities up to 98,5 % TD by pressureless sintering in various atmospheres. Full density is possible by post-HIP because the samples show only closed porosity. The microstructure was studied in dependence of sintering parameters. The level of impurities, i.e. C, O2 was determined. Hardness, fracture toughness and bending strength were measured for single phase and particle reinforced materials.

The primary and steady-state creep behavior of ductile-phase toughened Nb5Si3/Nb in-situ composites has been simulated using analytical and finite element (FE) continuum techniques. The microstructure of these composites is complex, consisting of large, elongated primary dendrites of the ductile (Nb) solid-solution phase in a eutectoid matrix with the silicide as the continuous phase. This microstructure has been idealized to facilitate the modeling; the effects of these idealizations on the predicted composite creep rates are discussed. Further, it has been assumed that the intrinsic creep behavior of each phase within the composite is the same as that of the corresponding bulk material. Thus, the experimentally measured creep properties of the bulk Nb5Si3 and (Nb) phases have been analyzed to provide the required material constants in the creep constitutive equation. Model predictions of the steady-state composite creep rate have been compared with the experimental results for a Nb-10 at.% Si alloy. While accurate at low stress, the models underpredict the composite creep rate at large stresses because the composite stress exponent is underpredicted. In the case of primary creep, the models somewhat over-predict the composite creep strain but are reasonably accurate given uncertainties in the primary creep data. Finally, FE predictions of the tensile stress distributions within the composites have been shown to be qualitatively consistent with the cracking observed experimentally during tertiary creep.

In-situ composites based on binary Nb-Si alloys and consisting of a Nb solid solution with Nb3Si or Nb5Si3 have shown a promising combination of low temperature and high temperature mechanical properties. The environmental resistance and room temperature fracture toughness of these composites can be further enhanced by additions such as Ti, Hf, Cr, and Al. In the present study, ternary Nb-Ti-Si alloys were prepared by directional solidification to generate aligned two and three phase composites containing a Nb solid solution with Nb3Si and/or Nb5Si3. The present paper will describe microstructures, phase equilibria and fracture toughness of these composites. The improvement in the room temperature fracture toughness over binary Nb-Nb5Si3 composites is discussed.

An investigation was undertaken to study the phase relations in Cr3Si alloyed with Mo varying from 10 to 83.5 wt. % of the material. Specimens were prepared from arc-melted buttons that were subsequently heat treated at 1673 K for 200 h and air quenched to room temperature to preserve the high temperature microstructures. Alloys containing more than 20 wt. % Mo were primarily two-phase materials of M3Si and M5Si3, where M is (Cr,Mo). Three alloys contained less than 5 % of a third phase, which also had the M5Si3 crystal structure. Differential thermal analysis (DTA) was performed on several specimens at temperatures up to 2073 K in order to determine a solidus curve for the M3Si phase. Since only one DTA peak was observed in each alloy, the M5Si3 phase must melt above 2073 K, the maximum temperature examined. A preliminary pseudo-binary phase diagram for (Cr,Mo)3Si and a portion of the 1673 K isothermal section of the Cr-Mo-Si ternary phase diagram are presented.

Previously, Cr-Cr3Si in-situ composites produced by arc-melting were shown to have good strengths at high temperatures1. For example, samples of a 25% Cr and 75% Cr3Si composite achieved bend strengths of 135 MPa at 1200°C. However, there is potential for even higher strengths at high temperatures and a need for improvement in the low temperature strength and toughness. In order to improve the properties, two approaches were taken. The first used powder metallurgy to develop a better microstructure than in the cast alloys, to try to improve both strength and toughness. The second approach was to incorporate erbia into the composites, to improve the strength and stability of the microstructure at elevated temperatures.

High density samples of hot pressed Cr-15.5Si and Cr-18.6Si have been produced by mixing Cr and Gr3Si powders and hot pressing in a graphite die. Erbia powders have been incorporated into some compacts for comparison.

Micros true tures have been characterized and mechanical properties determined. Both the hot pressing and the erbia affected the properties. In addition the erbia had a significant effect on consolidation of the samples.

Iron-rich Fe-Al alloys have been tensile tested in moist air and dry oxygen at a strain rate of 3.3×10−4 s−1. Moist air did not cause embrittlement until the composition reached 18-20% Al. Alloys with lower aluminum contents were embrittled when tested in hydrogen gas. The environmental sensitivity of these alloys was further investigated by examining the effects of strain rate on the ductility. For the most part, no significant strain rate effects were observed in the low aluminum alloys; strain rates of up to 3.3×10−1 s−1 were not fast enough to prevent embrittlement. In contrast, the ductility of Fe-35 at.% Al did increase with increasing strain rate in air and hydrogen; at a strain rate of 3.3×10−1 s−1 the elongations approached that of vacuum.

Structure, phase composition and air oxidation behavior in the temperature range 800 to 1400°C of a Nb-Ti-Al-based intermetallic alloy with the chemical composition (wt %): Nb-47.0; Ti-23.9; Al-21.0, V-4.4, and Cr-4.1 have been studied. The alloy structure is two-phase -σ (Nb2Al type) and γ (TiAl type). Preliminary air oxidation at 1400°C decreases the oxidation rate at 1150°C by a factor 2 to 3. It is connected with the formation of a protective scale (rutile and corundum with chromium and vanadium additions), refining of the alloy structure, and the formation of an internally oxidized underscale.

The corrosion behavior of a Ni3A1–1 5at%Fe alloy has been studied in air and melted salts, and also that of Ni3Al alloys contained series boron in melted salts. The first alloy has two oxidation rate constants. Its initial oxidation activity energy is 319 kJ / mol. There was much FeAl2O4 formed at 950°C, which possessed better oxidation resistance than NiFe2O4, therefore the alloy had the best oxidation resistance at 950°C. It also had better sulfidation resistance than 00Cr19NillTi stainless steel at 850°C, but the alloy was sulfidized seriously at 950°C. The sulfides were FeS2 at 850°C and A12S3, Ni2S3 at 950°C. There were different initial oxides formed, which were Fe3O4 at 850°C, and mixture oxides of NiO, NiMoO4, A12O3 and NiAl2O4 at 950°C , respectively. The Ni3Al alloys with boron contents from 0 to 3.7at% had better sulfidation resistance than the stainless steel at 850°C, and the alloy with 1.37at%B was the best. The sulfides were the same in the boron containing Ni3Al alloys which were Al2S3 and Ni7S6.

Ti-50(at.%)Al samples were oxidized at 900°C in air for 1 hour and studied by electron microscopy. The application of electron spectroscopic imaging with energy filtered TEM permitted the imaging of the distribution of the relevant elements (Ti, Al, O, N) within the oxide scale. With this method the formation of TiN in the inner part of the scale could be shown. The oxidation in air lead to an alternating scale containing TiN and alumina near the metal/scale interface. The formation of an aluminum-depleted second metallic phase in the metal beneath the oxide scale was observed.

Short-term cyclic oxidation tests were performed at 1200°C and 1500°C on ²-NiAl with 1 vol% oxide dispersions of Y2O3, La2O3 and Al2O3. The Y2O3 addition resulted in a very adherent ±-Al2O3 scale at 1200°C, with some improvement in the scale adhesion at 1500°C, compared to an undoped ²-NiAl alloy. The La2O3 and Al2O3 additions resulted in more severe oxidation than the undoped ²-NiAl at 1200°C. Both Y and La ions are found to segregate to ±-Al2O3 scale grain boundaries using scanning transmission electron microscopy/energy dispersive x-ray analysis. However, at both 1200°C and 1500°C, the La addition appears to cause a rapid increase in the isothermal oxidation rate, leading to excessive scale spallation.

Incipient melting and oxidation of grain boundaries at elevated temperatures have been investigated in the investment cast billet of a two-phase (γ + γ′) nickel aluminide alloy (Ni74.48Al16.98Cr8.02Zr0.51B0.10), designated as IC-218. Due to enrichment of Zr, which can form eutectic alloys with Ni, the inter-dendritic regions and grain boundaries melted incipiently at 1150°C. During quenching the molten layers often opened up producing cracks at the boundaries. Annealing at 1200°C in air for 570 min produced elongated ZrO2 precipitates at the grain boundaries. There was Zr depletion inside the grains in the vicinity of the precipitates. Such precipitates were scarcely observed in the specimens annealed in air at 1100°C for 23 h. Thus, the molten Ni-Zr alloy at the boundaries promoted internal oxidation.

In this study we present some results regarding phase and surface stability, with particular reference to the oxidation, of Ti alloys based on the γ-TiAl phase.

We investigated the stability of the metallic phases with TG and DSC methods, according to the temperature range over which the relevant transformations do occur. These tests were particularly useful to select times and temperatures for precipitation hardening treatments and also to evaluate the oxidation resistance.

The isothermal oxidation kinetics were followed with a thermobalance (TG) for time up to 25 hours and, for longer treatments, we put the samples in a furnace and measured their weight changes at fixed time intervals. The microstructure of the oxide scale was investigated with scanning electron microscopy and energy dispersion X-ray analysis and mapping. The identification of the crystalline phases was carried out with X-ray diffraction analysis.

The obtained results afford clear information on the predominant phenomena assisting the observed oxidation kinetics and give indications on the operating conditions suitable for these materials.

Self-consistent total energy calculations were performed to investigate the effects of ternary additions of 3d, 4d and 5d transition metal elements (V, Cr, Mn, Fe, Co, Ni, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Hf, Ta, W, Re, Os, Ir, Pt) to Ll0 ordered γ-TiAl. In general, we studied ordered solids of 4 atoms per unit cell. In the case of Mn additions larger supercells with up to 32 atoms were also treated. Minimizing the total energy with respect to volume and c/a ratio leads to the determination of lattice distortions due to alloying by the ternary elements. These distortions prove to be strongly dependent on whether the alloying element substitutes Al or Ti atoms. Furthermore, the site preference can be derived by comparing the formation energies of Tin-1XAln and TinXAln-i, where X is the substituting element. As a result the change of the geometry of L10 TiAl due to alloying can be estimated and may thereby give insight into the conditions at semicoherent γ/γ′ and α2/γ interfaces, which are associated to be responsible for ductility improvements of γ-TiAl based alloys.

The titanium-aluminum-erbium system has been studied using electromagnetic levitation and ultra high speed imaging in order to quantify the solidification velocity as a function of undercooling. Measurements were made on Ti-60 at% Al alloys with 0.25, 0.5, 1.0, and 2.0 at% Er additions. Boettinger and Aziz1 show that in ordered alloys it is thermodynamically possible at large enough interface velocities, corresponding to large undercoolings, to solidify a disordered phase with the same composition as the liquid. The result of the transition from ordered to disordered solidification is a discontinuity in the relationship between solidification velocity and melt undercooling. The experimental results for the solidification velocity as a function of melt undercooling will be presented and discussed for each alloy.

The mechanical properties of a β-containing Ti-Al-Cr alloy were investigated at ambient and elevated temperatures. The results show that the Ti-Al-Cr alloy containing the β phase has a very high tensile strength but a poor ductility at ambient temperature, and that higher ductility is obtained at high temperatures. The temperature dependence of mechanical properties is found to be sensitive to the strain rate during the test. Fractography shows that the fracture mode changes from fully brittle to ductile-brittle mixture with the increased temperature. All the results suggest that the triple-phased TiAl alloys (α2+β+γ) may have the combined mechanical properties of the dual-phased T13Al ((α2+β) and dual-phased TiAl (α2+γ) alloys.

The thermal stability of a TiAl based alloy was investigated through a series of high temperature exposures in air and vacuum. The microstructures were found to be thermally stable. The bending tests showed that both ductility and strength of air-exposed specimens decreased, but no significant variation was observed in those exposed in vacuum. The influence of alloy microstructure, thermal exposure temperature and time was also taken into account here.

A physically based micromechanical model is applied to study finite compressive deformation and failure in lamellar TiAl microstructures. Orientation dependent yielding of lamellar TiAl single crystals is specifically modeled. Finite element computations show that deformation is inherently non-uniform in the lamellar microstructure. Intergranular fracture (at low strains) and internal buckling (at high strains) are found to be two of the most important failure modes.

A continuous fibre reinforced SiC/TiAl composite has been prepared by arc spray and hot pressing. The as-fabricated fibre/matrix interface of monotapes and composites have been examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The arc-spray conditions result in a decohesion at the fibre/matrix interface.

The entire process of microcrack nucleation in a TiAl-Cr-V alloy with different microstructures is observed in situ by a scanning electronic microscope (SEM) during compressive deformation. The main results are summarized as follows: (1) In the fully lamellar microstructure, the microcracks nucleate mainly at the lamellar grain boundaries between soft-orientation and hard-orientation grains. (2) In the Widmanstatten structure, the microcracks distribute mainly at the intersection between two Widmanstatten plates. (3) In the duplex structure, the microcracks form at the lamella/ lamella and lamella/gamma boundaries. (4) In the nearly gamma structure, microcracks nucleate at gamma/gamma grain boundaries.