Richard W. Heckel

Kinetics of Phase Layer Growth During Aluminide Coating of Nickel.

The diffusion coating of nickel with aluminum was studied by a two-step aluminizing pack process involving initially an influx of aluminum at the surface (step 1) and later a partial honiogenization of the aluminum-rich region under conditions of zero surface flux (step 2). The process was studied in the temperature range from 870 to 1000°C. Step 1 was characterized mainly as the rapid, parabolic growth (after an initial transient period) of the Ni2Al3 phase (γ) as a surface layer with concurrent growth of a thinner NiAl (δ) layer. Step 2 was characterized mainly as the rapid loss of the aluminum gradient in the γ layer followed by parabolic growth of the δ layer primarily by the solution of they phase. Mathematical models were developed, in which numerical methods and computer techniques as well as closed-form solutions were utilized. The models yielded growth rate predictions in agreement with the experimental data and were used to define the critical parameters controlling growth kinetics for the aluminide layers formed during this process.

Solution kinetics of CuAl2 in an Al-4Cu alloy

The results of a recently-developed, spherical, finite geometry, mathematical model using numerical methods are compared to existing closed-form models. The geometry and boundary conditions imposed by the closed-form models are shown to be restrictive in the application of these models to the solution kinetics of second phases. A quantitative metallography study of the solution kinetics of CuAl2 in an Al-4 wt pct Cu alloy was undertaken to provide data on the changes in second-phase particle size distributions as a function of solution treatment time. The size distributions were shown to vary according to the proposed spherical, finite model formulated using numerical methods.

Transient growth of second phases during solution treatment

This study shows experimentally that, in the initial stages of a solution treatment process, an unstable phase may grow prior to dissolving if the diffusion flux in the unstable phase is larger than that in the stable matrix. The effect was demonstrated using diffusion couples fabricated from thin, multiple layers of α and β brass such that the mean compositions of the couples relative to the α-β phase equilibrium would result in solution of the β phase. Since at the diffusion temperature of 870°C the interdiffusion coefficient of the β phase is about twenty times that of the α phase, the initial β phase flux was much greater than that of the α phase. Metallographic observations of two sets of diffusion couples, each having a different mean composition, as a function of time of interdiffusion revealed that approximately 22 pct growth of the β phase preceded its ultimate dissolution. The analysis of this phenomenon was verified by reasonably good agreement between measured phase thickness values as a function of time and calculated values based upon a numerical solution of the appropriate multiphase diffusion problem.

The effects of heat treatment and deformation on the homogenization of compacts of blended powders

Homogenization of compacts of blended powders is reviewed in terms of the influence of variables which are known to affect the process. Mathematical models which describe major process variables are discussed and evaluated with experimental data. Consideration is given to the effects of both heat treatment and deformation on the rates of homogenization in one-phase and multiphase binary systems. Guidelines are presented which will assist other workers in the use of the mathematical models as predictive tools.

Optimization of the range of elastic behavior of unidirectional composites by prestraining

The effect of tensile prestrain on the stage I tensile yield stress has been studied both analytically and experimentally for composites whose stress-strain curves obey the rule of mixtures. The mathematical analysis provides a means for calculating the optimum amount of prestrain, the residual stresses (in the direction of the fibers) in the matrix and fiber materials after unloading from the prestraining, and the stage I yield stress in the composite after the prestrain treatment. It is shown that the improvement in stage I yield stress by prestraining is due to the development of negative residual stresses in the matrix. The stage I yield stress in composites with negligible residual stresses in the as-fabricated condition can usually be improved by a factor of two by prestraining; the amount of improvement is even greater if the as-fabricated composites have the usual state of residual stress,i.e., tension in the matrix. Experimental studies on 2024 aluminum-tungsten composites (filament-wound; hot-pressed) having tungsten fiber volume fractions between 0.08 and 0.40 verified the mathematical analysis. The stage I yield stresses measured in these composites after a prestrain of 4.2 × 10−3 were in good agreement with predicted values. Improvements of up to a factor of six were found in the stage I yield stress as a result of prestraining.

Homogenization of compacted blends of Ni and Mo powders

The homogenization behavior of compacted blends of Ni and Mo powders was studied primarily as a function of temperature, mean compact composition, and Mo powder particle size. All compact compositions were in the Ni-rich terminal solid-solution range; temperatures were between 950 and 1200°C (in the region of the phase diagram where only the MoNi intermediate phase forms); average Mo particle sizes ranged from 8.4 μm to 48 μm. Homogenization was characterized in terms of the rate of decrease of the amounts of the Mo-rich terminal solid-solution phase and the MoNi intermediate phase. The experimental results were compared to predictions based upon the three-phase, concentric-sphere homogenization model. In general, agreement between experimental data and model predictions was fairly good for high-temperature treatments and for compact compositions which were not close to the solubility limit of Mo in Ni. Departures from the model are discussed in terms of surface diffusion contributions to homogenization and non-uniform mixing effects.

Homogenization of compacted blends of nickel and tungsten powders

The diffusional homogenization of compacted blends of elemental Ni and W powders was studied using quantitative metallography and X-ray compositional line broadening. Homogenization treatments were carried out in the temperature range from 1156 to 1271°C (in the region of the Ni-W phase diagram where only the terminal solid solutions exist). The mean compositions of the compacts studied ranged from 0.06 to 0.14 atomic fraction tungsten (in the Ni-rich solid solution). Other variables studied included the size and shape of the W particles, compaction pressure, and the effect of applied load during the homogenization treatment. Experimental measurements of the solution of the unstable W-rich phase and the increase in W concentration in the Ni-rich matrix were compared to calculated values obtained from a concentric-sphere mathematical model which assumed volume interdiffusion (concentration-independent diffusion coefficients) and local equilibrium at the heterophase interface. The model provided a fairly good approximation of the homogenization kinetics and the effects of the major processing and powder parameters investigated. The primary sources of discrepancy between model predictions and experimental data appeared to result from modest surface diffusion contributions to homogenization and imperfect powder blending.

Fabrication of composites by diffusion processing

Abstract : Composite materials containing continuous NiAl3 fibers in an aluminum matrix were grown by solid state diffusion at 600C from nickel wires embedded in aluminum (hot-pressed). The kinetics of formation of the NiAl3 and Ni2Al3 phases and the ultimate solid of the nickel terminal solid solution and Ni2Al3 phases were studied. The NiAl3-aluminum composite structure was developed by a diffusion treatment of 72 hours at 600C when 0.005 inch nickel wires were used in the starting structure. Although strengths of these composites were low due to grain boundary fracture of the NiAl3, this method of composite fabrication may prove applicable to the fabrication of composites in other alloy systems. (Author)

A mathematical model study of the influence of degree of mixing and powder particle size variation on the homogenization kinetics of compacted blends of powders

The rates of diffusional homogenization of compacted blends of powders are sometimes less than predictions based upon mathematical diffusion models. Discrepancies between model calculations and experimental data are most apparent during the later stages of homogenization and are generally ascribed to the effects of nonuniform particle mixing and variations in particle size which are not treated by the models. The present mathematical model study (using both numerical and analytical techniques) was undertaken as a means of estimating the magnitudes of the effects of mixing and particle size variations on homogenization rates. A one-dimensional model (“stacked plates”) was employed wherein two solute-rich plates of the same size or different sizes could be positioned at various locations within a distance element. The remaining spaces were assumed to be filled with solvent-rich plates. An isomorphous alloy system was considered. Thus, the model was a one-dimensional, two-plate analog of the distribution of solute-rich particles in a continuous matrix of very fine solvent-rich particles. The effects of variations in solute-rich plate positions (degree of mixing) and of unequal solute-rich plate thicknesses (size distributions) were both shown to influence homogenization significantly. It is concluded that the variations in the local environments of the solute-rich particles in a compacted powder blend can result in decreases in the rate of diffusional homogenization that are of the magnitude of the departures of actual experimental data from predictions of models which assume uniform mixing of solute-rich particles of constant size.