Numerical results have been obtained for the plastic flow of a metal matrix composite material reinforced by rigid spheres, cylinders and ellipsoids. Aligned geometries have been investigated and 2-d and 3-d cellular calculations carried out using the finite element method. The relative effect of particulates, disks and needles used as reinforcements are compared. Needles in aligned arrangements are found to produce the greatest strengthening. Additional results are presented from plane strain slip line solutions. Such results provide insight into the relative importance to strength of the reinforcement end regions versus the narrow side ligaments.

Investigation of the flow properties of materials containing rigid reinforcements has been prompted by various technological challenges concerned with high specific strength composites. The materials of interest range from metallic materials such as Al alloys, to ceramic materials such as Si3N4 and also include the intermetallics: TiAl and MoSi2. The reinforcements are typically A12O3 and SiC. The strengths of interest may be the flow strength at ambient temperature or the creep strength. The present analysis constitutes an attempt to provide estimates of either the flow or creep strength for matrices that exhibit power law deformation.

A model based on the micromechanical mechanism of crack growth resistance in fiber reinforced brittle matrix composites is presented. The formulation of the model is based on a small scale geometry of a macrocrack with a bridging zone, in this case the process zone, which governs the resistance mechanism. The effect of high toughness of the fibers in retardation of the crack advance, and the significance of the fiber pullout mechanism on the crack growth resistance, are reflected in this model. The model allows one to address issues such as influence of the fiber spacing and fiber-matrix friction.

Two specific cases are analyzed. One represents the fracture initiation and concentrates on the development of the first microcrack between fibers. The second case deals with the development of an array of microcracks forming the bridging zone. In both cases a discrete fiber distribution is assumed. The analysis is based on an exact solution of the corresponding. boundary value problem.

Although intermetallics based upon aluminum or silicon tend to have a very attractive combination of low density and excellent oxidation resistance, they suffer from lack of adequate creep strength and, in most cases, from inadequate ductility and toughness. It has been recognized for several years that an approach which could simultaneously solve both problems, without degrading other properties, is to utilize the intermetallics as matrices for composite materials. The consequence has been an explosion of interest in two-phase intermetallicbased alloys, as is manifested in the current symposium.

One method of imparting damage tolerance into intermetallic compounds is to introduce ductile particles in a fine uniform distribution. Systematic criteria are given as to the necessary requirements for such ductile phase reinforced systems. Two composite systems, Cr2Nb/Nb and Nb3Al/Nb were selected for further evaluation. Both microstructural evaluations as well as thermal stability were demonstrated in both systems. Small cracks introduced via micro-hardness indentations were measured as a function of precipitate size. In Cr2Nb/Nb the crack lengths decreased with increasing particle size which was explained through increasing effectiveness of ductile particles with minimum effective particle radii calculated at 6-10 μm. Larger particles along with poor ductile phase distributions led to lower damage tolerance in aged Nb3Al/Nb. By maintaining uniform ductile particle distributions, intermetallic compounds possessing superior damage tolerance have been demonstrated.

Pressureless reaction sintering was investigated as a fabrication method for composites composed of a MoSi 2 based matrix with up to 20 vol% SiC reinforcement. Up to 10 vol% of a number of reactants including Mo+2Si, CrSi2, and Cr+2Si were utilized. In nearly all cases the reactants greatly enhanced pure matrix densification at temperatures as low as 1400°C. The composites containing SiC reinforcement did not densify nearly as well. For some reactants it appears that an adverse reaction occurred between the SiC and the additives resulting in swelling.

The addition of titanium diboride to near gamma titanium aluminides has been effective in reducing the initial grain size formed on solidification from the melt. Despite this refinement, however, significant constitutional inhomogeneity may be found due to segregation of alloying elements and reinforcement phases during solidification. The dendritic solidification of several XD'T alloys was studied to assess the extent of this segregation. Quantitative elemental analysis of dendritic and interdendritic regions indicates that a substantial chemical gradient is developed in aluminum and ternary alloying elements. The extent of chemical and reinforcement segregation and its relationship to overall aluminum content in the alloys is discussed. In a study to determine thermomechanical processing steps required to eliminate the casting inhomogeneity, homogenization heat treatments and isothermal forging were performed and the effect on microstructure determined. Remnant dendritic cell structures were seen to persist despite high-temperature heat treatments and mechanical working.

The feasibility of producing layered composite sheet material, in which the matrix is the intermetallic compound TiAl, using physical vapor deposition (PVD) has been demonstrated. The PVD techniques of sputtering and electron beam evaporation were both investigated. Films were deposited by alternately sputtering from targets of composition Ti-53A1–3Nb and TiB2 or were co-deposited from separate electron beam heated hearths containing Ti and Al. In the latter case, nitrogen gas was pulsed into the deposition chamber at controlled intervals resulting in the formation of a TiAl/TiAlN layered composite. The composition of the films was determined using Auger analysis and Rutherford backscattering spectroscopy and the crystal structure was checked using X-ray and electron diffraction. Cross-sections through the films were prepared for examination by transmission electron microscopy so that the layered microstructure of the films could be verified.

The feasibility of fabricating Nb-Nb silicide microstructure by Physical Vapor Deposition with sufficient control of impurities has been investigated. It is demonstrated that electron beam evaporation can satisfy the requirement of impurity control under appropriate vacuum condition. In elemental layered structure of Nb and Si, NbSi2 is the first phase formed upon annealing. It was found that, at 600°C, the growth rate of NbSi2 phase at the interface of Nb and Si is 35Å/min. The Nb5Si3 phase nucleates at a higher temperature (around 900°C) at the interface of Nb and NbSi2. In the case of co-deposited film with overall composition around Nb5 Si3, NbSi2 formation is by-passed. Thus, multilayers of Nb/NbSi2 or Nb/Nb5Si3 can be formed from layered elemental deposition and subsequent heat treatment under controlled conditions by adjusting the starting thicknesses of the films. Alternate elemental and co-deposition and subsequent or in-situ heat treatment can directly form the layered Nb/desired silicide composite. Multilayers of Nb/NbSi2 and Nb/Nb5Si3 with layer thicknesses below 500Å have been formed from layered elemental deposition.

NbA13 matrix composites were synthesized “in-situ” via reactive hot compaction (RHC) of elemental powders. It was found that the simultaneous application of pressure during synthesis was effective in attaining a near-theoretical density matrix at relatively low temperatures and pressures. Using this technique, two types of composites were produced: (1) matrices containing a uniform dispersion of second phase particles (either Nb3Al or Nb2Al with an Nb core or Nb2Al) and (2) matrices reinforced with coated or uncoated ductile Nb filaments. It was found that a limited amount of toughening is obtained using the first approach, while composites containing coated Nb filaments exhibited a significant increase in the ambient temperature fracture toughness. In this paper, various aspects of RHC processing of NbAI3 matrix composites, the effect of initial stoichiometry and powder size on the microstructure, as well as the mechanical behavior of the composites will be discussed.

Additions of ∼5%B1 and ∼9%Ta to binary γ-TiAl result in the formation of a monoboride isomorphous with TiB but containing Ta in solid solution. This boride exhibits strongly anisotropic growth from the melt, producing rod-like primary phases which are of potential interest for creep strengthening of the γ matrix. Fibrous borides with aspect ratios larger than ∼20 and volume fractions of ∼0.12 have been produced in arc-melted Ti-48AI-9Ta-4.3B alloys. Creep testing at 1255 K indicates that these “in-situ” composites have much higher creep strength than γ-TiAl, but the contribution of the reinforcements is relatively small compared with the solute-strengthening of the matrix.

The floating zone directional solidification technique has been applied to an XDTM short-fiber-reinforced titanium aluminide ingot in an effort to produce in situ alignment of the reinforcement. Microstructural evaluation reveals that a general alignment of the high aspect ratio (>100:1) fibers occurs under specific imposed solidification conditions. These metallographic observations are supported by 800 ° C mechanical data, which indicate higher axial and reduced transverse strengths relative to the unprocessed base material which contains a dispersion of randomly oriented fibers. The increased strengths are observed to be a consequence of an increase in the matrix hardening due to accommodation of plastic strain around the fiber reinforcement.

Microstructural development in alloys bordering the NbCr2 Laves phase composition has been evaluated following solidification processing. With splat quenching of binary alloys, rapid solidification has yielded supersaturation of both Nb-rich and Cr-rich primary solid solutions as well as some solubility extension of the C15 phase during cooling as compared to as-cast structures. Ternary additions of Ti are soluble in NbCr2 along the pseudobinary NbCr2-TiCr2 section for C15 structures and completely miscible in bcc Nb and Cr solutions. In the rapid freezing of splats, modest Ti additions to NbCr2 enhance the development of an extended bcc solution during freezing and the supersaturation of the C15 phase during solid-state cooling. In the solid state, the C14–C15 transformation of the Laves phase is suppressed partially in arc-cast ingots and more completely at high cooling rates. Upon heating, thermal analysis has revealed that the retained C14 phase undergoes a decomposition sequence with at least two stages, possibly involving polytypes and associated faulting, to the stable C15 structure. The microstructural options developed by solidification processing, solid state transformations, and alloying have been observed to modify the brittle behavior of the as-cast structures.

Fusion zone microstructures of an electron beam (EB) welded XDt m Ti–48at%Al + 6.5 vol% TiB2 alloy revealed plate-like precipitates which were absent in the base metal. The volume fraction of this phase increased with increasing cooling rate and correlated with increased weld cracking frequency. To determine whether this phase was a product of solidification from the melt or a product of a solid-state transformation, the microstructures of the welds were compared to those of samples cycled in a Gleeble 1500/20 Thermal-Mechanical Test System which was programmed to simulate the solid-state portion of the weld cooling rates (as predicted by a Rosenthal analysis). The microstructures were characterized by X-ray diffraction, optical and by scanning electron microscopy. The plate-like phase found in the weld microstructures was identified as TiB2 occurring upon rapid solidification of the melted weld metal.

MoSi2 based composites possess significant potential as high temperature structural materials. The present investigation constitutes an initial study of the fabrication and properties of MoSi2/silicide alloys and SiC particle reinforced-MoSi2 /silicide alloy matrix composites. Alloying disilicides (WSi2, Nb5Si2, Ta5Si2 ) and trisilicides (Mo5 Si3, W5Si3, Nb5Si3, Ta5Si3, Ti5Si3 ) were examined. Results indicated that at the 50/50 mole % alloying level, MoSi2 formed solid solutions with the alloying disilicides. Reaction of the MoSi2 /TaSi2 alloy with SiC was observed. An initial assessment of 50/50 mole % disilicide alloy oxidation behavior suggested an oxidation resistance inferior to that of pure MoSi2.

MoSi2 matrix composites have been recognized lately as potential materials for structural applications at elevated temperatures. Specifically, MoSi2 composites may exhibit useful properties at temperatures to 1400°C. Previous work improved the yield strength of MoSi2 at 1400°C by a factor of five through SiC whisker reinforcement. Current research is directed towards increasing the fracture toughness of MoSi2 through the addition of ductile phase reinforcements such as niobium and tantalum. The reaction between Nb and MoSi2 to form (Mo,Nb)5 Si3 proceeds with faster kinetics at hot isostatic press temperatures as low as 1100°C when compared to the reaction between Ta and MoSi2 to form (Mo,Ta)5Si3. This reaction product exhibits very poor properties, as evidenced by crack propagation through this layer during fracture. The feasibility of hot working these composites to produce tailored microstructures is examined.

MoSi2-based composites reinforced with particles, whiskers and continuous fibers are fabricated using hot pressing and hot isostatic pressing techniques. The microstructure, interface compatibility, mechanical properties and failure behavior of the resulting composites are discussed.

The quasibinary NiAI-Mo system exhibits a large two-phase field between NiAl and the terminal (Mo) solid solution, and offers the potential for producing in-situ eutectic composites for high-temperature structural applications. The phase stability of this composite system was experimentally evaluated, following long-term exposures at elevated temperatures. Bend strengths as a function of temperature and room-temperature fracture toughness data are presented for selected NiA1-Mo alloys, together with results from fractography observations.

The lack of low-temperature damage tolerance along with the desire to achieve lower densities and higher creep strengths over monolithic intermetallic materials, has sparked significant interest in the research community for the development of an intermetallic matrix composite suitable for high temperature structural applications. Eutectic systems are desirable as composites due to the inherent thermodynamic stability between matrix and reinforcement. It also allows the opportunity of controlling the microstructure by control of solidification conditions. This paper presents a qualitative examination of the toughness behavior of the Cr-Cr2Zr, Cr-Cr2Hf, Cr-Cr2Ta, Cr-Cr3Si, Ta-Cr2Ta, and Ta-Ta5Si eutectics. The evaluation is based on microcracking from microhardness indentations and fracture appearance. Relative oxidation resistance at 800°C and 1200°C is also presented.

Hot extrusion of premixed charges of 10–20 vol% chopped A12O fiber and Ni3Al powder has resulted in composite alloys of near theoretical density. In tensile tests at room temperature, density-compensated yield strengths of some of these composite alloys are as good or better than those of as-cast Ni3Al without reinforcement; however strengths at 1000°C in vacuum are lower. The low strength at 1000°C and fine grain size (2–3/μm) suggest the presence of superplastic behavior and the accompanying diffusional creep and grain boundary sliding. This presentation discusses our findings to date and includes microstructural studies and tensile properties, both at room temperature in air and 1000°C in vacuum.