Search

The TiBe12/NiAl composites fabricated by a powder metallurgy technique were examined for the extent of interdiffusion between its components, using scanning Auger microscopy. The results revealed an evidence of a reaction between the two component intermetallics after annealing at 1300 K for 100 h, with a 95-micron reaction zone extending from the TiBe12/NiAl interface. The concentration profile across the reaction zone showed that the Ni, Al, and Be concentrations were in a 1:1:2 ratio, while Ti was present in only very small amounts.

A Co-free modified superalloy similar in composition to Waspaloy is investigated in an effort to understand the effect of Co on reaction zone growth kinetics and verify the chemistry dependence of reaction zone growth in the matrix of tungsten fiber reinforced superalloy composites. The values of the parabolic rate constant, characterizing the kinetics of reaction zone growth, for the Waspaloy matrix and the C-free alloy as well as five other alloys from a previous study confirm the dependence of reaction zone growth kinetics on cobalt content of the matrix. The Co-free alloy composite exhibits the slowest reaction zone growth among all tungsten fiber reinforced composites studied to date.

Interdiffusion and interdiffusion controlled intermediate phase formation in metal matrix composites can be of interest for the prolonged application of these systems at high temperatures. Methods are discussed that address the kinetics of interdiffusion in systems that exhibit solid solution intermixing or the formation of a third intermediate phase at the fiber/matrix interface, or both. The tungsten fiber reinforced niobium and tungsten fiber reinforced superalloy systems are employed as model systems for experimentation and discussion. In an effort to impede interdiffusion, the concept of ion implanted diffusion barriers have been examined. Preliminary results on the feasibility and effectiveness of ion implanted diffusion barriers are presented.

While cast polycrystalline superalloys recommend themselves in virtue of better 'buy-to-fly' ratios and higher strengthening gamma-prime volume fractions than those of wrought superalloys, the expansion of their use into such critical superalloy applications as gas turbine hot section components has been slowed by insufficient casting process opportunities for microstructural control. Attention is presently drawn, however, to casting process developments facilitating the production of defect-tolerant superalloy castings having improved fracture reliability. Integrally bladed turbine wheel and thin-walled turbine exhaust case near-net-shape castings have been produced by these means.

Transient model of monolith catalytic combustor presented in report done under NASA/DOE contract. Model assumes quasi-steady gas phase and thermally "thin" solid. In gas-phase treatment, several quasi-global chemical reactions assumed capable of describing CO and unburnt hydrocarbon emissions in fuel-lean operations. In steady-state computation presented, influence of selected operating and design parameters on minimum combustor length studied. When fast transient responses required, both steady and unsteady studies made to achieve meaningful compromise in design.

A stainless steel, two-dimensional (rectangular), center-dump, premixed-prevaporized combustor with quartz window sidewalls for visual access was designed, built, and used to study flashback. A parametric study revealed that the flashback equivalence ratio decreased slightly as the inlet air temperature increased. It also indicated that the average premixer velocity and premixer wall temperature were not governing parameters of flashback. The steady-state velocity balance concept as the flashback mechanism was not supported. From visual observation several stages of burning were identified. High speed photography verified upstream flame propagation with the leading edge of the flame front near the premixer wall. Combustion instabilities (spontaneous pressure oscillations) were discovered during combustion at the dump plane and during flashback. The pressure oscillation frequency ranged from 40 to 80 Hz. The peak-to-peak amplitude (up to 1.4 psi) increased as the fuel/air equivalence ratio was increased attaining a maximum value just before flashback. The amplitude suddenly decreased when the flame stabilized in the premixer. The pressure oscillations were large enough to cause a local flow reversal. A simple test using ceramic fiber tufts indicated flow reversals existed at the premixer exit during flickering. It is suspected that flashback occurs through the premixer wall boundary layer flow reversal caused by combustion instability. A theoretical analysis of periodic flow in the premixing channel has been made. The theory supports the flow reversal mechanism.

A stainless steel, rectangular center-dump, premixed-prevaporized combustor was used at the NASA Lewis Research Center to study flashback in controlled conditions simulating gas turbine combustor operations under steady-state and transient conditions. The combustor allowed visual access to the combustion area, and was operated at inlet air temperatures from 600 to 850 K, premixer wall temperatures from 450 to 900 K, and average premixer velocity from 40 to 80 ft/s. A parametric study revealed a slight decrease in the flashback equivalence ratio as the inlet air temperatures increased. It also indicated that the premixer wall temperature and premixer velocity are not governing parameters of flashback. It is suggested that flashback occurs through the premixer wall boundary layer flow reversal caused by combustion instability.

A theoretical model is proposed for the determination of phase equilibrium in alloys, taking into consideration dissimilar lattice parameters. Volume-dependent pair interactions are introduced by means of phenomenological Lennard-Jones potentials and the configurational entropy of the system is treated in the tetrahedron approximation of the cluster variation method. The model is applied to the superalloy-relevant, nickel-rich, gamma/gamma-prime phase region of the Ni-Al phase diagram. The model predicts reasonable values for the lattice parameters and the enthalpy of formation as a function of composition, and the calculated phase diagram closely approximates the experimental diagram.

Flashback, a problem that occurs in premixed-prevaporized combustors, is the upstream propagation of the flame from the combustor into the premixing tubes. Not only does flashback change the combustion process from premixed burning to diffusion burning, thus creating more pollutants, but it also inflicts considerable damage to the fuel injector, premixing tube and other equipment upstream. The conditions at which flashback occurs in steady burning and the mechanism that causes flashback in both steady and transient flow are studied. The equivalence ratio at which flashback occurs is being measured for inlet temperatures of 600-950 K, premixer wall temperatures of 450-1050 K and premixer velocities of 40-80 ft/s. These data are presented.

Cobalt was systematically replaced with nickel in Waspaloy (which normally contains 13% Co) to determine the effects of cobalt on the creep behavior of this alloy. Effects of cobalt were found to be minimal on tensile strengths and microstructure. The creep resistance and the stress rupture resistance determined in the range from 704 to 760 C (1300 to 1400 C) were found to decrease as cobalt was removed from the standard alloy at all stresses and temperatures. Roughly a ten-fold drop in rupture life and a corresponding increase in minimum creep rate were found under all test conditions. Both the apparent creep activation energy and the matrix contribution to creep resistance were found to increase with cobalt. These creep effects are attributed to cobalt lowering the stacking fault energy of the alloy matrix. The creep resistance loss due to the removal of cobalt is shown to be restored by slightly increasing the gamma' volume fraction. Results are compared to a previous study on Udimet 700, a higher strength, higher gamma' volume fraction alloy with similar phase chemistry, in which cobalt did not affect creep resistance. An explanation for this difference in behavior based on interparticle spacing and cross-slip is presented.

The effect of cobalt on the workability of nickel-base superalloys is examined with reference to experimental results for four heats of alloys based on the Nimonic 115 composition with varying amounts of nickel substituted for the nominal 14 percent cobalt. It is shown that Co lowers the gamma-prime solvus, which in turn lowers the Cr23C6 carbide solvus. It is further shown that these solvus temperatures bracket the hot working range for the alloys. However, thermomechanical processing modifications reflecting the effect of Co on the gamma-prime and carbide solvi are shown to restore the workability and the properties of alloys with little or no cobalt.

The role of cobalt in a representative wrought nickel-base superalloy was determined. The results show cobalt affecting the solubility of elements in the gamma matrix, resulting in enhanced gamma' volume fraction, in the stabilization of MC-type carbides, and in the stabilization of sigma phase. In the particular alloy studied, these microstructural and microchemistry changes are insufficient in extent to impact on tensile strength, yield strength, and in the ductilities. Depending on the heat treatment, creep and stress rupture resistance can be cobalt sensitive. In the coarse grain, fully solutioned and aged condition, all of the alloy's 17% cobalt can be replaced by nickel without deleteriously affecting this resistance. In the fine grain, partially solutioned and aged condition, this resistance is deleteriously affected only when one-half or more of the initial cobalt content is removed. The structure and property results are discussed with respect to existing theories and with respect to other recent and earlier findings on the impact of cobalt, if any, on the performance of nickel-base superalloys.

A typical innovation-to-commercialization process for the development of a new hot section gas turbine material requires one to two decades with attendant costs in the tens of millions of dollars. This transfer process is examined to determine the potential rate-controlling steps for introduction of future low strategic metal content alloys or processes. Case studies are used to highlight the barriers to commercialization as well as to identify the means by which these barriers can be surmounted. The opportunities for continuing joint government-university-industry partnerships in planning and conducting strategic materials R&D programs are also discussed.

The effect of cobalt or substituting for cobalt on the mechanical properties of nickel-based superalloys is discussed. Waspaloy, UDIMET 700, and NIMONIC 115, which are representative of Ni-Cr-Co-Al-Ti-Mo superalloys having different gamma prime contents which are strengthened by a heavily alloyed matrix, coherent gamma prime precipitates, and carbides at the grain boundaries. Microstructure and in situ and extracted phase STEM micro-analysis were used to evaluate the three alloys.

Development of a predictive method for determination of the gamma/gamma prime phase fields, i.e., gamma prime volume fraction as a function of the multicomponent composition, is described. The cluster variation method used for binary alloys in which the precipitated phase is coherent with the matrix phase is extended for application to the multicomponent coherent gamma/gamma prime nickel-based superalloys. It is shown that the cluster variation method can accurately describe the equilibrium (incoherent) gamma/gamma prime phase fields in the binary Ni-Al phase diagram. The gamma/gamma prime phase field for the Ni-Cr-Al ternary phase diagram is computed as a function of temperature. A reasonable fit results between the calculated and the experimental diagrams. The modeling of the six-component Ni-Cr-Al-Co-Mo-Ti base superalloy and the effect of Ni substitution of Co are discussed.

The creep and stress rupture properties of an oxide (Y2O3) dispersion strengthened nickel-base alloy, which also is strengthened by gamma-prime precipitates, was studied at 760 and 1093 C. At both temperatures, the alloy YDNiCrAl exhibits unusually high stress rupture ductility as measured by both elongation and reduction in area. Failure was transgranular, and different modes of failure were observed including crystallographic fracture at intermediate temperatures and tearing or necking almost to a chisel point at higher temperatures. While the rupture ductility was high, the creep strength of the alloy was low relative to conventional gamma prime strengthened superalloys in the intermediate temperature range and to ODS alloys in the higher temperature range. These findings are discussed with respect to the alloy composition; the strengthening oxide phases, which are inhomogeneously dispersed; the grain morphology, which is coarse and elongated and exhibits many included grains; and the second phase inclusion particles occurring at grain boundaries and in the matrix. The creep properties, in particular the high stress dependencies and high creep activation energies measured, are discussed with respect to the resisting stress model of creep in particle strengthened alloys.

The effect of cobalt on the basic mechanical properties and microstructure of wrought nickel-base superalloys has been investigated experimentally by systematically replacing cobalt by nickel in Udimet 700 (17 wt% Co) commonly used in gas turbine (jet engine) applications. It is shown that the room temperature tensile yield strength and tensile strength only slightly decrease in fine-grained (disk) alloys and are basically unaffected in coarse-grained (blading) alloys as cobalt is removed. Creep and stress rupture resistances at 760 C are found to be unaffected by cobalt level in the blading alloys and decrease sharply only when the cobalt level is reduced below 8 vol% in the disk alloys. The effect of cobalt is explained in terms of gamma prime strengthening kinetics.

A study has been carried out to assess the role of cobalt in Udimet 700, a representative nickel-base superalloy containing 17 percent or more cobalt. The study spans the spectrum of microstructural, microchemical, and mechanical behavior aspects which together form a basis for superalloy performance in jet engines. The results suggest that cobalt affects the solubility of elements in the gamma matrix, which leads to enhanced gamma-prime volume fraction and to the stabilization of MC-type carbides and sigma phase. However, these microstructural and microchemical changes are too slight to significantly affect the strength and ductile properties. Depending on the heat treatment, the creep and stress rupture resistance can be cobalt-sensitive. In the coarse-grained, fully solutioned and aged condition, all of the alloy's 17 percent Co can be replaced by nickel without decreasing the creep and stress rupture resistance. These findings are discussed with reference to existing theories and experimental data obtained by other workers.

A quasi-steady gas phase and thermally thin substrate model is used to analyze the transient behavior of catalytic monolith combustors in fuel lean operation. The combustor response delay is due to the substrate thermal inertia. Fast response is favored by thin substrate, short catalytic bed length, high combustor inlet and final temperatures, and small gas channel diameters. The calculated gas and substrate temperature time history at different axial positions provides an understanding of how the catalytic combustor responds to an upstream condition change. The computed results also suggest that the gas residence times in the catalytic bed in the after bed space are correlatable with the nondimensional combustor response time. The model also performs steady state combustion calculations; and the computed steady state emission characteristics show agreement with available experimental data in the range of parameters covered. A catalytic combustor design for automotive gas turbine engine which has reasonably fast response ( 1 second) and can satisfy the emission goals in an acceptable total combustor length is possible.

High-temperature structural applications usually require creep resistance because some average stress is maintained for prolonged periods. Alloy and microstructural design guidelines for creep resistance are presented through established knowledge on creep behavior and its functional dependences on alloy microstructure. Important considerations related to creep resistance of alloys as well as those that are harmful to high-temperature properties are examined. Although most of the creep models do not predict observed creep behavior quantitatively, they are sophisticated enough to provide alloy or microstructural design guidelines. It is shown that creep-resistant microstructures are usually in conflict with microstructures that improve such other properties as stress rupture ductility. Greater understanding of the effects of environments on creep and stress rupture behavior of materials is necessary before one can optimally design alloys for applications in different environments.

Nickel-base, cobalt-base, and high nickel and chromium iron-base alloys are dissected, and their microstructural and chemical components are assessed with respect to the various functions expected of high temperature structural materials. These functions include the maintenance of mechanical integrity over the strain-rate spectrum from creep resistance through fatigue crack growth resistance, and such alloy stability expectations as microstructural coarsening resistance, phase instability resistance and oxidation and corrosion resistance. Special attention will be given to the perennial conflict and trade-off between strength, ductility and corrosion and oxidation resistance. The newest developments in the constitution of high temperature alloys will also be discussed, including aspects relating to materials conservation.

Prolonged high temperature performance of structural alloys requires joint attention to both corrosion and mechanical properties, and to their possible interactions. In this interpretive review paper, we provide a systematic compaction of theories and key observations on corrosion-mechanical properties interactions, from early single crystal work in the 1930s to recent studies on superalloys in nonoxidizing, oxidizing and hot corrosive environments. Particular attention is paid to environmental effects on creep, stress rupture and fatigue properties.

It is pointed out that the major difficulty concerning an understanding of the flaw growth resistance properties of superalloys is related to the lack of basic theories. An attempt is made to develop these theories, taking into account their use for the identification of tentative alloy design guidelines for flaw growth resistance. A brief review of the fatigue crack growth literature is provided. Attention is given to continuous fatigue crack growth theory, threshold fatigue crack growth, planar slip superalloy fatigue crack growth, and aspects of creep crack growth resistance.

The effect of elastic modulus and the temperature dependence of elastic modulus on creep activation energies for an oxide dispersion strengthened nickel-base superalloy are investigated. This superalloy is commercially known as Inconel Alloy MA 753, strengthened both by gamma-prime precipitates and by yttria particles. It is shown that at intermediate temperatures, say below 1500 F, where elastic modulus is weakly dependent on temperature, the modulus correction term to creep activation energy is small. Accordingly, modulus corrections are insignificant for the superalloy considered, which shows high apparent creep activation energies at this temperature. On the contrary, at very high temperatures, the elastic modulus correction term can be significant, thus reducing the creep activation energy to that of vacancy self-diffusion. In order to obtain high-temperature creep resistance, a high-value elastic modulus with a weak dependence on temperature is required.

The strengthening microstructural features in heat-resistant super-alloys are oxide dispersoids and/or gamma-prime precipitates. The role of the strength of the alloy matrix in the creep resistance of these alloys is considered. An evaluation was made of experimental data on oxide-dispersion- and/or precipitation-strengthened nickel-based alloys with various levels of matrix solid solution strengthening; a generalized expression for creep rates which separates the matrix contributions from the particle contributions to the resisting stress and creep strength of these alloys was derived. It is concluded that the major role of the alloy matrix in the creep behavior of these alloys is in determining the apparent stress dependence of the creep rates.

A steady-state lean combustion model for monolithic catalytic combustors is given. The model, consisting of several semi-global chemical reaction steps in the gas-phase and on the surface, is capable of analyzing CO and THC emissions. In the model computation presented, the influence of operating and design parameters on the minimum combustor length is studied. Special attention is given to the effect of after-bed gas-phase reaction space. Comparison with experimental data indicates good agreement in the range of parameters covered.

Effect of the length available for gas-phase reactions downstream of the catalytic reactor on the emission of CO and unburned hydrocarbons was investigated. A premixed, prevaporized propane/air feed to a 12/cm/diameter catalytic/reactor test section was used. The catalytic reactor was made of four 2.5 cm long monolithic catalyst elements. Four water cooled gas sampling probes were located at positions between 0 and 22 cm downstream of the catalytic reactor. Measurements of unburned hydrocarbon, CO, and CO2 were made. Tests were performed with an inlet air temperature of 800 K, a reference velocity of 10 m/s, pressures of 3 and 600,000 Pa, and fuel air equivalence ratios of 0.14 to 0.24. For very lean mixtures, hydrocarbon emissions were high and CO continued to be formed downstream of the catalytic reactor. At the highest equivalence ratios tested, hydrocarbon levels were much lower and CO was oxidized to CO2 in the gas phase downstream. To achieve acceptable emissions, a downstream region several times longer than the catalytic reactor could be required.

Recent developments in the interpolation of high-temperature steady-state creep results have introduced the concept that the stress dependence of the creep rate should be in terms of the effective stress referred to as the applied stress minus a back stress. This paper reports on back stresses taken from data on a gamma-prime-strengthened wrought nickel-base superalloy, an oxide dispersion-strengthened ODS nickel-base solid solution alloy, and an ODS nickel-base superalloy. The effect of air versus vacuum environments and the effect of dynamic changes in the strengthening microstructures on the magnitude of the back stress are assessed. The role of modulus normalization and the back stress correction in determining the true creep activation energy are examined. It is shown that the high values of the apparent stress exponent 'n' of the steady-state creep equation can be easily explained through a relationship between n, the true stress exponent of steady-state creep, and the stress which when subtracted from the applied stress results in the effective driving stress acting on the mobile dislocations during creep.

The Larson-Miller (L-M) method of extrapolating stress rupture and creep results is based on the contention that the absolute temperature-compensated time function should have a unique value for a given material. This value should depend only on the applied stress level. The L-M method has been found satisfactory in the case of many steels and superalloys. The derivation of the L-M relation is discussed, taking into account a power law creep relationship considered by Dorn (1965) and Barrett et al. (1964), a correlation expression reported by Garofalo et al. (1961), and relations concerning the constant C. Attention is given to a verification of the validity of the considered derivation with the aid of suitable materials.

The mechanical behavior of particle-strengthened alloys is reviewed. On the basis of this knowledge, it is concluded that second-phase particles, coherent or incoherent, can enhance the flow strength, creep resistance, and stress-rupture life of alloys. Unfortunately, particles are usually not beneficial alloy-design elements if enhanced uniaxial ductility, plane-strain ductility, stress-rupture ductility, and toughness are called for. Such properties as fatigue-crack propagation resistance appear to require, for example, both high strength and high ductility, a situation which can come to pass only when the perennial conflict between strength and ductility is resolved in particle-strengthened systems in particular and in any other material system in general. Wherever possible, the role of coherent and incoherent particles in alloy design is distinguished.

The paper derives an analytical expression for creep crack growth rate based on a model in which the higher and concentrated stresses ahead of the crack tip enhance creep deformation, thus progressively causing stress rupture and continuous crack advance. The equation derived for creep crack growth rate is expressed in terms of a geometrical factor in the stress intensity expression and in terms of the crack tip radius whose lower limit estimate is the fracture mechanics crack-tip opening displacement. The functional features of the derived creep crack growth rate equation can be compared with those of a pertinent empirical equation. The analytical expression provides some guidelines for alloy and microstructure design.

Control and causes of combustion instability in rocket engines