Interdiffusion and Growth of the Phases in CoNi/Mo and CoNi/W Systems

Deleterious topological-closed-packed (tcp) phases grow in the interdiffusion zone in turbine blades mainly because of the addition of refractory elements such as Mo and W in the Ni- and Co-based superalloys. CoNi/Mo and CoNi/W diffusion couples are prepared to understand the growth mechanism of the phases in the interdiffusion zone. Instead of determining the main and cross-interdiffusion coefficients following the conventional method, we preferred to determine the average effective interdiffusion coefficients of two elements after fixing the composition of one element more or less the same in the interdiffusion zone. These parameters can be directly related to the growth kinetics of the phases and shed light on the atomic mechanism of diffusion. In both systems, the diffusion rate of elements and the phase layer thickness increased because of the addition of Ni in the solid solution phase, probably because of an increase in driving force. On the other hand, the growth rate of the μ phase and the diffusion coefficient of the species decreased because of the addition of Ni. This indicates the change in defect concentration, which assists diffusion. Further, we revisited the previously published Co-Ni-Mo and Co-Ni-W ternary phase diagrams and compared them with the composition range of the phases developed in the interdiffusion zone. Different composition ranges of the tcp phases are found, and corrected phase diagrams are shown. The outcome of this study will help to optimize the concentration of elements in superalloys to control the growth of the tcp phases.     

Microstructural investigation of intermediate phase formation in uranium-aluminum diffusion couples

The irregular growth of the intermediate phase layer which forms as a result of interdiffusion between U and Al has been studied. Using optical and scanning electron microscope examination techniques and electron-probe microanalysis, it has been determined that only one phase (UAl₃) grows in the U-Al system and that the breakdown of interfacial planarity is a result of a limited nucleation rate and rapid growth rate of UA1₃. This phase partially decomposes near the uranium end of the diffusion couple during slow cooling, resulting in the formation of a “multiphase≓ region. No such region is observed in rapidly cooled specimens. X-ray diffraction indicates the absence of a preferential growth direction to the intermediate phase layer. A model is presented that describes the sequence of intermediate phase growth.     

Diffusion in β2 Fe-Ni-Al alloys

Diffusion in the β(bcc) phase field of the Fe-Ni-Al system was investigated at 1004°C with solid-solid diffusion couples assembled with β₂ alloys of selected composition. Experimental diffusion paths were determined for all couples and interdiffusion coefficients calculated at composition points corresponding to intersections of diffusion paths and maxima and minima of concentration profiles. The dependence of interdiffusion coefficients on composition was most clearly presented in terms of the parameter Fe/(Fe + Ni). The diffusive interactions between aluminum and nickel as represented by the cross coefficients were either positive or negative depending on the ternary composition. The Fe/(Fe + Ni) ratio appeared to be a significant parameter since iron and nickel atoms behave differently in affecting the degree of ordering in nonstoichiometric (Fe, Ni)Al alloys with less than 50 at. pct aluminum.     

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. R. W. HECKEL, formerly Professor of Metallurgical Engineering, Drexel University, Philadelphia, Pa. A. J. HICKL, formerly Graduate Student, Drexel University     

An analysis of interdiffusion in finite-geometry,two-phase diffusion couples in the Ni−W and Ag−Cu systems

The progress of homogenization in finite, multilayer diffusion couples was studied experimentally using both quantitative metallography to measure phase thicknesses and electron microprobe analysis to determine concentration-distance profiles. Ni?W couples having mean compositions in the nickel-rich terminal solid solution range (12.1 and 15.2 at. pct W) were studied after 4.43 to 240.7 hr interdiffusion treatments at 1156° and 1207°C. Ag?Cu couples having mean compositions in both of the terminal solid solution ranges (8.5 at. pct Cu and 2.1 at. pct Ag) were studied after 3.8 to 65.4 hr interdiffusion treatments at 760°C. Experimental data were in good agreement with calculations of interdiffusion based on equilibrium interface concentrations and concentration-independent interdiffusion coefficients. The good agreement between experimental data and calculated values for the Ni?W couples extended to times necessary for the achievement of essentially complete homogenization,i.e., through both the two-phase dissolution process and the subsequent stage of gradient elimination in the terminal solid solution phase.     

Interdiffusion in the Fe-Pt System

Diffusion-couple experiments are conducted in the Fe-Pt system. The phase boundary compositions of the phases measured in this study are found to be different than the compositions published previously. In the γ-FePt solid solution, the interdiffusion coefficient increases with the Pt content up to 25 at. pct Pt. Fe is the faster diffusing species in this phase. The trend in the interdiffusion coefficient is explained with the help of calculated driving force for diffusion. To reduce errors, the average interdiffusion coefficients are calculated in the FePt and FePt₃ compounds.     

Reaction diffusion in the NiCrAl and CoCrAl systems

The oxidation life of a physically vapor-deposited overlay protective coating based on the Ni-Cr-Al or Co-Cr-Al (M-Cr-Al) systems is controlled by aluminum consumption resulting from alumina spalling, erosion, and interdiffusion with the substrate. The rates of these processes are determined by coating and substrate composition, service environment, and temperature. The purposes of this study were 1) to determine the effect of coating and substrate chemistry on the kinetics of interdiffusion and their relation to diffusion-zone constitution and 2) to develop a procedure for analyzing diffusion in multicomponent, multiphase systems. Semi-infinite diffusion couples with MCrAl sources representative of coatings and sinks representative of nickel-or cobalt-base gas turbine alloys were annealed at 1000, 1095, 1150, or 1205°C for as long as 500 h. The couples were examined by optical microscopy and layer growth measurements were made. Parabolic growth constants can be correlated with source aluminum content for a specific sink composition since changes in source aluminum content are more important than changes in chromium content. Sink composition is as important as source composition in determining β-recession kinetics in diffusion couples and life in finite coatings. Nickel-base alloys are more stable sources and stronger sinks than cobalt-base alloys. Total and diffusion activation energies were determined from layer growth constants. By introducing the concept of β source strength which can be determined from the appropriate phase diagrams, the Wagner solution for consumption of a second phase in a semiinfinite couple was successfully applied to the analysis of the MCrAl couples. This provided a method 1) for correlating β-recession rate constants with couple composition, 2) for determining reliable total and diffusional activation energies, and 3) for calculating interdiffusion coefficients.     

The early stage of Ni3AI layer growth in NiAI/Ni diffusion couples

The growth kinetics and morphology of the Ni₃Al intermediate phase present in NiAI/Ni diffusion couples have been studied for short times ranging from a few minutes, or less, to a few hours at 1100 °C. Despite high heating rates (<240 seconds to reach 1100 °C), a significant portion of the Ni₃Al layer forms upon heating such that the layer growth which occurs during heating is approximately equal to that which occurs during the first hour of isothermal interdiffusion. TEM studies indicate that the irregular interfaces present at both interphase boundaries are associated with grain boundaries within the Ni₃Al phase. A qualitative model accounting for enhanced layer growth by a grain boundary contribution to diffusion within the Ni₃Al layer is described. formerly with the Department of Metallurgy and Materials Engineering, Lehigh University, Bethlehem, PA,.     

High temperature reaction zone growth in tungsten fiber reinforced superalloy composites: Part I. Application of the moving boundary equations

The nature, kinetics, and rate-controlling process for high temperature reaction zone growth in tungsten fiber reinforced superalloy (TFRS) composites are determined. The matrix alloys considered in this study are INCOLOY 903 and a W-modified INCOLOY 903. The kinetics and rate-controlling process for reaction zone growth are determined by direct application of the moving boundary equations. The results show that reaction zone growth is a results of longrange interdiffusion and that the rate-controlling process for reaction zone growth is interdiffusion across the reaction zone. Further, the results show that the interdiffusion coefficients of the reaction zone can be calculated from the parabolic rate constants for the reaction zone growth without directly considering the kinetics of diffusion in the matrix. The activation energy,Q, and the frequency factor,D _o, for the interdiffusion across the reaction zone in the INCOLOY 903 matrix composite are experimentally determined to be 268 kJ/mole and 0.22 cm²/s, respectively. Further, the relative chemical stability of the reaction zone is discussed with respect to reaction zone chemistry. T. CAULFIELD, formerly Postdoctoral Research Scientist at Columbia University, is Senior Member of Research Staff, Philips Laboratories, Briarcliff Manor, NY 10510. J. K. TIEN is Henry Marion Howe Professor and Director, Center for Strategic Materials, Henry Krumb School of Mines, Columbia University, New York, NY 10027.     

Analysis of low-temperature intermetallic growth in copper-tin diffusion couples

A multiphase diffusion model was constructed and used to analyze the growth of the ε- and η-phase intermetallic layers at a plane Cu-Sn interface in a semi-infinite diffusion couple. Experimental measurements of intermetallic layer growth were used to compute the interdiffusivities in theε andη phases and the positions of the interfaces as a function of time. The results suggest that interdiffusion in the ε phase(≈D _ε) is well fit by an Arrhenius expression with D₀ = 5.48 × 10⁻⁹ m²/s andQ = 61.9 kJ/mole, while that in the η phase (≈D_η) has D₀ = 1.84 × 10⁻⁹ m²/s andQ = 53.9 kJ/mole. These values are in reasonable numerical agreement with previous results. The higher interdiffusivity in theη phase has the consequence that theη phase predominates in the intermetallic bilayer. However, the lower activation energy for interdiffusion in theη phase has the result that theε phase fills an increasing fraction of the intermetallic layer at higher temperature: at 20 °C, the predicted ε-phase thickness is ≈10 pct of that ofη, while at 200 °C, its thickness is 66 pct of that ofη. In the absence of a strong Kirkendall effect, the original Cu-Sn interface is located within theη-phase layer after diffusion. It lies near the midpoint of theη-phase layer at higher temperature (220 °C) and, hence, appears to shift toward the Sn side of the couple. The results are compared to experimental observations on intermetallic growth at solder-Cu interfaces.     

Modelling the influences of solid-state interdiffusion and dissolution on transient liquid phase sintering kinetics in a binary isomorphous system

A simple model was developed to predict the impact that solid-state interdiffusion and dissolution have on liquid formation and its duration during transient liquid phase sintering (TLPS). The model predicts that solid-state interdiffusion can dramatically reduce the amount of liquid initially formed during heating. This reduction is dependent on the heating rate and initial base metal particle size. In cases of sintering above the additive phase melting point, the model predicts that base metal dissolution increases liquid phase formation and that this additional melting reduces the base metal particle size. The model predicts that longer times are required to solidify isothermally the greater amounts of liquid formed at higher temperatures (because of dissolution). This agreed qualitatively with experimental results for a Ni-65 wt pct Cu TLPS mixture sintered at 1090 °C and 1140 °C. Quantitative comparisons between the model and experiment were good at 1140 °C; however, the rate of isothermal solidification was underestimated by the model for intermediate sintering times at 1085 °C.     

Magnetic contribution to the interdiffusion coefficients in bcc (<Emphasis Type="Italic">α</Emphasis>) and fcc (<Emphasis Type="Italic">γ</Emphasis>) Fe-Ni alloys

The interdiffusion coefficients in bcc (α) and fcc (γ) Fe-Ni alloys below their Curie temperatures have been calculated based on the magnetic contribution to the free energy for interdiffusion. The free energy for interdiffusion due to magnetic ordering in bcc Fe-Ni alloys is positive. The calculated interdiffusion coefficients in bcc Fe-Ni alloys fit the experimental data quite well. In fcc Fe-Ni alloys, the magnetic contribution to interdiffusion depends on both temperature and composition and is abnormal for Ni compositions in the Invar region. The free energy of vacancy formation is positive and the free energy of vacancy migration is negative, due to the effect of magnetic ordering. The interdiffusion coefficient in the ferromagnetic phase is lower than that extrapolated from the paramagnetic phase for Ni compositions of 50 at. pct and greater and is higher than that extrapolated from the paramagnetic phase for Ni compositions of 40 at. pct and lower.     

Experimental and numerical study of pattern formation in faceted cellular array growth

In this article, a review of our recent experimental and numerical study of pattern formation in faceted cellular array growth was presented.In situ observations in transparent faceting organic compounds have revealed the dynamical features of faceted cellular growth. Cell tip splitting and loss of cells in the array have been found to be the main mechanisms for cell spacing adjustment. The time evolution of a faceted cellular array has been followed numerically. It was found that pattern formation was history-dependent and was determined by interactions in the array. These interactions were either transient or persistent, depending on the growth condition. As a result, afinite range of stable cell spacings was found under a given growth condition; that is, there was no sharp selection. The cellular structures were irregular when persistent interactions occurred, whereas relatively regular structures could be formed once the transient interactions had stopped. The implication of these observations for pattern formation in other array growth processes, such as nonfaceted cellular, dendritic, or eutectic growth, will be discussed. This paper is based on a presentation made in the symposium “Growth and Configurations of Faceted Solid-Liquid Interfaces” presented during the TMS annual meeting, New Orleans, Louisiana, February 17–21, 1991, under the auspices of the TMS Solidification Committee. Formerly Assistant Research Engineer, The University of Alabama.     

Evolution of interaction domain microstructure during spray deposition

An interaction domain, defined in the present article as the region where semisolid, atomized droplets impinge and are collected during spray atomization and deposition, was systematically investigated on the basis of a semisolid metal-forming mechanism. Accordingly, microstructural evolution in the interaction domain was rationalized by quantitative analyses of (1) the solid fraction of semisolid metal, (2) the extent of deformation and deformation strain rate, and (3) the solidification cooling rate. The results demonstrate that the fraction of solid in the interaction domain ranges from 0.5 to 0.8 for the materials studied here: Ni₃Al, Al-6 wt pct Si, and Al-6 wt pct Fe. Moreover, the results show that the semisolid material in the interaction domain experiences a severe deformation during deposition with an associated strain rate of up to 10⁶ s^-1. As a result of this deformation; the solidification structure is modified, and, in particular, any dendritic structure that is present will undergo extensive fragmentation. The severe deformation that is experienced by the interaction domain and the presence of a solidification cooling rate that is on the order of 10 to 10₅ Ks^-1 were proposed to be critical factors that promote the formation of a spheroidal grain morphology during spray atomization and deposition. Experimental support to this suggestion was provided by microstructural observations on Ni₃Al, Al-6Si, and Al-6Fe. In particular, the morphological modification of the primary Si phase that is observed in spray-atomized and spray-deposited Al-6Si was rationalized on the basis of these factors.     

Application of computational thermodynamics to Fe/Ni,Fe-3%Si/Ni and 316L/Ni systems produced by powder metallurgy

The commercial simulation software packages, Thermo-Calc and DICTRA, are widely used to predict phase formation and diffusion in materials prepared by traditional metallurgical processes. The goal of this work was to clarify whether the software can also be applied to materials produced by powder metallurgy techniques. For this purpose, diffusion couples of the material combinations Fe/Ni, Fe-3%Si/Ni and 316L/Ni were prepared by two techniques: co-pressing of powders followed by co-sintering and joining of two sintered parts followed by annealing. Thermo-Calc and DICTRA were used to predict interface formation and phase transformations during heat treatments. Scanning electron micrography and energy-dispersive X-ray spectrometry showed the formation of interdiffusion interfaces. The diffusion rate was higher in co-sintered couples, resulting in broader interdiffusion zones than those predicted by simulations, but the annealed couples exhibited interdiffusion profiles similar to those predicted by DICTRA simulations. In general, simulations and experimental results showed the same tendencies.     

Damage propagation in multidirectional composites subjected to compressive loading

The mode of compressive damage growth in multidirectional composites is investigated. A detailed systematic study of the nature of damage evolution is presented and the microstructural basis for treating compressive damage as a cracklike feature is explained. Experimental and theoretical findings conclude that there is a strong scientific basis for treating compressive damage growth in fiber-reinforced laminated polymer matrix composites as a bridged mode-I crack in compression. A large-scale crack-bridging model has been successfully used to predict the steady-state damage propagation stress as a function of damage growth in notched composites. These predictions are in good agreement with experimental observations.     

Growth of silicides and interdiffusion in the Mo-Si system

Solid-solid diffusion couples assembled with disks of Mo and Si were annealed at selected temperatures, over the temperature range from 900 °C to 1350 °C, for the development of diffusion structure and determination of interdiffusion coefficients for the silicides of Mo. Layers of MoSi₂ and Mo₅Si₃ were observed to form in the diffusion zone; the MoSi₂ layer was one to two orders of magnitude larger in thickness than the Mo₅Si₃ layer. The MoSo₂ layer developed a columnar microstructure with evidence of texture and preferential growth of grains in the direction of diffusion. The Si-to-Mo ratio, determined by microprobe analysis across the thickness of the MoSi₂ layer, varied within the approximate range from 1.9 to 2.0. From the concentration profiles, integrated interdiffusion coefficients as well as energies of activation for interdiffusion were determined for the silicide layers. On the basis of the observed stoichiometric range for the MoSi₂ phase, average values of the interdiffusion coefficients were also estimated. Relations are derived between the growth-rate constant and the integrated interdiffusion coefficient for the MoSi₂ phase. The evaluated activation energies for interdiffusion in the MoSi₂ and Mo₅Si₃ phases are 130±20 and 210±10 kJ/mol, respectively.     

In Situ Investigation of the Allotropic Transformation in Iron

Previous studies on the allotropic transformation in iron suggested a high transformation temperature and a controversial transformation mechanism. At present, however, our understanding of the transformation characteristics in iron seems insufficient, especially as there is a general lack of in situ data on morphological evolution and lattice parameter changes during the transformation. In this study, the characteristics of the morphology, the crystallographic relationships, and short‐range diffusion in the allotropic transformation are investigated systematically on the basis of in situ and ex situ observations using time resolved X‐ray diffraction, laser scanning confocal microscopy, and electron backscattering diffraction. In contrast to the martensitic transformation, the characteristics of the allotropic transformation will be highlighted. Finally, the results are compared with previous reports in order to provide further insight into the nature of the phase transformation based on the in situ observations.     

A new analysis for the determination of ternary interdiffusion coefficients from a single diffusion couple

Concentration profiles that develop in a ternary diffusion couple during an isothermal anneal can be analyzed directly for average ternary interdiffusion coefficients. A new analysis is presented for the determination of average values for the main and cross-interdiffusion coefficients over selected regions in the diffusion zone from an integration of interdiffusion fluxes, which are calculated directly from experimental concentration profiles. The analysis is applied to selected isothermal diffusion couples investigated with α (fcc) Cu-Ni-Zn alloys at 775 °C, β (bcc) Fe-Ni-Al alloys at 1000 °C, and γ (fcc) Ni-Cr-Al alloys at 1100 °C. Average ternary interdiffusion coefficients treated as constants are calculated over composition ranges on either side of the Matano plane and examined for the diffusional interactions among the diffusing components. The ternary interdiffusion coefficients determined from the new analysis are observed to be consistent with those determined by the Boltzmann-Matano analysis at selected compositions in the diffusion zone. The ternary interdiffusion coefficients are also employed in analytical solutions based on error functions for the generation of concentration profiles for the selected diffusion couples. The generated profiles are a good representation of the experimental profiles including those exhibiting uphill diffusion or zero-flux plane (ZFP) development for the individual components. Uncertainties in the values of the interdiffusion coefficients calculated on the basis of the new analysis are found to be minimal.