G. R. Edwards

Modeling of infiltration kinetics for liquid metal processing of composites

An equation for modeling the kinetics of liquid-metal infiltration into a porous compact has been developed. The model is based on considering a bundle of capillary tubes as an analog for the porous compact. A solution which describes a limiting form of behavior has been shown to be valid for small extents of infiltration relative to a hypothetical static state. The numerical solutions of the dimensionless infiltration-equation have been used to delineate conditions for which limiting solutions are valid. A dimensionless group λ has been shown to be capable of classifying the behavior into two limiting cases: either “inviscid-flow“, λ 10-2, or “viscous-flow“, λ > 102. It would appear that for capillary-tube (pore) radii less than 100 µ-m and for conditions where the compact is wetted by the liquid metal, A is likely to be >100 and therefore correspond to a “viscous-flow system“. Also, an infiltration-rate parameter, ϕ, has been selected which can be used to assess the effects of alloying additions to the liquid-metal infiltrant. This parameter can therefore provide for the selection of alloy infiltrants, surface preparation modification, and processing parameters such as temperature, time of infiltration, and pore size of compact in regard to the processing of composites. Although the model concept (capillary-tube bundle) is recognized as being deficient in not treating the compact as what it really is (a porous medium), the work presented was intended to quantify the limiting behavior of the capillary-tube-bundle approach which has been used in the past.

A thermodynamic criterion to predict wettability at metal- alumina interfaces

Metals are known to wet ceramics by chemical bond formation. Existing theories, using reaction thermodynamics, can predict the relative wetting trends in wetting systems but fail to distinguish between the wetting and non wetting systems. Wetting is considered as a surface phenomenon, and the spontaneity of wetting is controlled by ΔGW, a thermodynamic term defined here as the Gibbs free energy of wetting. A model that treats wetting as a reaction between the surface phase of the ceramic and the molten metal is presented to calculate ΔGW. The model is used to predict wetting tendencies of various molten metals on α-alumina surfaces. The predictions are compared with previously published results, as well as with the experimental results of this study. Experimental wettability parameters were measured using a capillary rise apparatus. Measurements were made for various metals wetting an aluminum-oxide surface. Based on this model, a thermodynamic wetting map that delineates wetting and non wetting regimes is drawn. A map of this nature can be used advantageously in ceramic joining and metal-matrix composites applications.

Heat treatment of investment cast PH 13-8 Mo stainless steel: Part I. Mechanical properties and microstructure

The microstructure of investment cast PH 13-8 Mo stainless steel heat-treated to various conditions was studied using light and electron microscopy, electron probe microanalysis, and Mössbauer spectroscopy. The mechanical properties were investigated by using uniaxial tensile testing, hardness testing, and Charpy impact testing. TheΒ-NiAl strengthening precipitates, though detectable by electron diffraction, were difficult to resolve by transmission electron microscopy (TEM) in specimens aged at low temperatures (566 °C and below). A high dislocation density was observed in the lath martensitic structure. The higher strength and lower ductility observed at low aging temperatures was attributed to both the high dislocation density and the precipitation ofΒ-NiAl. When samples were aged at high temperatures (> 566 °C), a lower dislocation density and a reverted austenite fraction on the order of 15 pct were observed. SphericalΒ-NiAl precipitates were observed in the overaged condition. The decrease in strength and corresponding increase in ductility observed in samples aged at temperatures above 566 °C were attributed to the reverted austenite and recovery. Mechanical properties were improved when the homogenizing temperature and time were increased. Electron probe microanalysis quantified the increased homogeneity realized by increasing homogenizing temperature and time. Elimination of the refrigeration step, which normally follows the solution treatment, did not degrade the mechanical properties. Mössbauer spectroscopy showed only minor decreases in the fraction of retained austenite when refrigeration followed the solution treatment.

The infiltration of aluminum into silicon carbide compacts

Although liquid-metal processing of metal matrix composites offers economic advantages, problems related to the nonwetting nature of the ceramic discontinuous reinforcement create obstacles to its ready implementation. Infiltration can occur only if a threshold pressure is applied to overcome the unfavorable interfacial forces in the system. The research reported in this paper has been devoted primarily to experiments on infiltrating silicon carbide compacts with pure aluminum, aluminum-1 wt pet magnesium, and aluminum-1 wt pet silicon. The major finding has been that an incubation time is necessary before infiltration can proceed, even though the threshold pressure is exceeded. Thus, while the model equations available for predicting the infiltration rate of compacts appear to be adequate, the incubation time can represent the rate-determining step in the process. It is suggested that the mechanism responsible for the incubation phenomenon may be related to a surface modification produced by either reaction of liquid aluminum with an oxide film on the surface of the particles or coverage of the surface by a capillarity-induced aluminum condensate.

Factors controlling the magnesium weld morphology in deep penetration welding by a CO2 laser

In laser welding with power density beyond 104 W · mm−2, the formation of plasma cavities, commonly referred to as keyholes, leads to deep penetration welds with high aspect ratios. In this paper, the morphologies of keyhole welds produced with a 6 kW CW CO2 laser on two die-cast magnesium alloys, AZ91 and AM50, are compared. It was found that the two magnesium alloys responded differently to laser welding. Though irregular weld cross-section profiles were consistently observed on each materials, bead dimensions often varied with the welding variables in contrasting ways. For both alloys, important characteristics of the weld beads such as depth, width, crown height (hump), and surface ripples were analyzed as a function of the welding parameters, most particularly the heat input. Results show that the use of heat input, a variable grouping two welding parameters into one, was often inadequate in characterizing the bead morphology. Several explanations are given, including base metal vaporization, but the process of bremsstralung absorption explains it well and rationalizes many observed characteristics of laser weld morphology.

Heat treatment of investment cast PH 13-8 Mo stainless steel: Part II. Isothermal aging kinetics

The hardening response of investment cast PH 13-8 Mo stainless steel has been evaluated by hardness measurements following aging in the temperature range normally specified for this alloy (510 °C to 593 °C). A new relationship between fraction transformed and hardness was developed, and analysis of the data in terms of the kinetics of precipitation, in a manner similar to that frequently applied to other precipitation-hardenable martensitic steels, yielded low time exponents and a low value for the apparent activation energy. The values of the time exponents were 0.49, 0.37, 0.56, and 0.53 at 510 °C, 538 °C, 566 °C, and 593 °C, respectively, and that for the apparent activation energy was 139 kJ/mole. As has been proposed for other maraging type steels, these estimates suggest thatΒ-NiAl precipitates along or near dislocations and that growth of the precipitates is dominated by dislocation pipe diffusion. However, these predictions were neither supported nor refuted by transmission electron microscopy (TEM) because of difficulties in imaging theΒ-NiAl precipitates at the aging times and temperatures used. Further, analysis of the data using the formalism of Wert and Zener for the growth of precipitates with interfering diffusion fields indicated that the estimates of fraction transformed from hardness data are not fully appropriate for maraging type steels. Consideration of the nature of the Avrami analysis and the electron microscopy results suggests that other phenomena, including dislocation recovery and reversion of martensite to austenite, occur at rates sufficient to convolute the Avrami analysis. It is further suggested that these results cast doubt on the fundamental implications of previous analyses of precipitation kinetics in age-hardening martensitic steels. Although the Avrami analysis was found not to provide a tenable description of the precipitation kinetics, it does provide a reasonable methodology for portrayal of the hardening response of PH 13-8 Mo stainless steel.

Kinetics of interlayer formation on polycrystallineα-Al2O3/copper-titanium alloy interface

Copper-titanium alloys are known to wet an alumina surface. Traditionally, contact angles are used as measurable parameters to monitor the progress of wetting. Contact angle is a measure of the thermodynamic equilibrium at the interface. For reactive wetting systems, it is well known that the interfacial properties vary as a function of time. In the present study, kinetics of reaction were studied by monitoring the rate of interfacial phase formation. An immersion apparatus was built for this purpose. The extent of reaction was measured both by a simple surface compositional analysis and by interfacial reaction layer thickness measurements. The reaction layer exhibited a parabolic growth with an associated activation energy of 180 to 230 kJ/mole. A speculative growth mechanism is proposed based on the experimental observations and the information available in the literature.

Low-temperature solid-state bonding of copper

Mechanisms associated with low-temperature solid-state bonding of copper were investigated. The low-temperature bonding phenomena was observed to be time/temperature dependent and related to a deformation process. An analytical technique for predicting bond strength and bonding behavior is presented.

Development of an electronic phase diagram and the predictions of plutonium alloy phase stability using electronic properties

Experimentally measuring electronic properties could have a significant impact in determining the kinetics of aged plutonium alloys. Phase stability of plutonium alloys can be assessed by using modified empirical electronic models in conjunction with measurements of electronic and magnetic properties of plutonium alloys. The issues surrounding the evaluation of aged plutonium alloys and the evolution of electronic-based alloy theories as applied to the prediction of the solubility of gallium in plutonium are presented. Using solid solution thermodynamics in combination with these electronic models, the phase diagram for dilute solid solutions can be estimated. There are numerous measurements that could be used to determine the relationship between a materials phase stability and its electronic structure. The measured properties of interest here are the Seebeck coefficient (thermopower), Hall coefficient and electrical resistivity. Combining three property measurements (or three other phase sensitive properties) into a three-dimensional plot with a specific property on each axis, a region of space will be formed that describes the stability of the phase. Also, the implications of assessing electronic data with this methodology are discussed.

Effect of sea water on the fatigue crack propagation characteristics of welds for offshore structures

The fatigue crack propagation characteristics of ASTM A36 steel weldments produced by both conventional surface welding operations and underwater repair welding techniques were evaluated in both air and seawater. The resulting fatigue crack growth rates were shown to depend sensitively on pore density and pore distribution, factors that varied significantly with welding procedure and environment. Variations in growth rate were correlated with scanning electron fractographs of the stable fatigue crack zone. The results of this study show that underwater wet welding procedures produce fatigue-resistant weld metal that is adequate for use at low applied stresses in offshore structures.