Non-isothermal finite diffusion-controlled growth in ternary systems

A mathematical model has been developed for diffusion controlled phase growth in ternary systems. Local equilibrium at phase boundaries and one dimensional diffusion controlled growth is assumed. The model includes a method of determining phase growth velocity and interface compositions consistent with the diffusion rate of both solute elements. This method also accounts for the effects of overlapping diffusion fields and nonisothermal growth. Initial conditions can be any curvilinear composition gradients and boundary conditions can be fixed or vary with time and/or temperature. The Crank-Nicolson finite difference equations are used to provide numerical stability and flexibility. Other capabilities of the model include treatment of finite systems, of nonisothermal phase growth and of off-diagonal ternary coefficients (D ₂₁ ³,D ₁₂ ³). Several sample simulations of the constant cooling of a 2.1 wt pct P, 4.1 wt pct Ni, 93.8 wt pct Fe alloy are presented. Three cooling rates are used: 5×10⁻³, 5×l0⁻⁴, and 5×l0⁻⁵ °C/s. An Fe-Ni-P alloy of this same composition was cooled in the laboratory for five days at 5×lo⁻⁴ °C/s from 900 to 685°C. Excellent agreement was found for the predicted and measured composition gradients and precipitate sizes.     

Ternary diffusion in the α and γ phases of the Fe-Ni-P system

Ternary diffusion coefficients have been determined in the α-bcc and γ-fcc phases of the Fe-Ni-P system and at four temperatures 1200, 1100, 1000, and 900°C. At 1100°C the ratio ofD _NiNi ^Fe /D _PP ^Fe in the α phase is 0.3 to 0.4 and the ratio of the cross coefficientD _PNi ^Fe /D _PP ^Fe is 0.03 to 0.05. m the γ phase the corresponding ratios are 0.01 to 0,02 and 0.0015 to 0.0025. The other cross coefficientD _NiP ^Fe could not be evaluated because of experimental uncertainties. Estimates of the ratioD _NiP ^Fe /D _NiNi ^Fe using the interaction parameter ε₁₂ are 0.004 to 0.01 in the α phase and 0.02 to 0.04 in the γ phase. The addition of P in both the α and γ phase increases the major ternary coefficients up to as much as a factor of ten at one temperature. This is consistent with the fact that P lowers the melting point of FeNi in the ternary system, up to 500°C. Isodiffusion coefficient contours obtained at 1100°C plot approximately parallel to the α and γ solidus boundaries and are similar in shape to the α and γ solidus boundaries as a function of temperature. Activation energies andD ₀ values were computed at selected compositions in both α and γ phases forD _PP ^Fe andD _NiNi ^Fe and are given below: Formerly Graduate Student, Department of Metallurgy and Materials Science, Lehigh University, Bethlehem, Pennsylvania     

The structures expected in a simple ternary eutectic system: Part II. The Al-Ag-Cu ternary system

Ternary alloys of various compositions from the aluminum rich corner of the Al-Ag-Cu system were directionally solidified at several different growth rates ranging from 6.4 × 10⁻¹ mm·S⁻¹ to 5.6 × 10⁻³ mm· s⁻¹. The region of two phase coupled growth between α-Al and CuAl₂ was determined at a growth rate of 6.4 × 10⁻¹ mm· s⁻¹. The composition range over which a fully ternary eutectic structure formed was investigated for several different growth rates. The results are found to be consistent with the predictions of the competitive growth model set out in Part I,¹ and it would seem that the ternary eutectic composition of the published phase diagram may be incorrect. Scanning electron microscopy, using the backscattered electron signal, was used, together with optical microscopy, to study the microstructures formed. The ternary eutectic between α-Al, Ag₂Al, and CuAl₂ was found to be semiregular, and the unusual morphology of the two phase dendrites between α-Al and Ag₂Al is explained. Formerly with Alcan Laboratories Ltd., Banbury, Oxon, United Kingdom Formerly with Alcan Laboratories Ltd.     

Ternary dissolution kinetics in the Fe-Ni-P system

The dissolution of (FeNi)₃P in the ternary Fe-Ni-P system has been studied by optical and electron microprobe techniques. Precipitates of (FeNi)₃P, initially in equilibrium with their ternary matrix (a at 750°C, y at 875°C), were examined after being partially dissolved by heating at 975°C. In addition, diffusion couples with starting compositions similar to the equilibrated ternary alloys were examined after also being heat treated at 975°C. Phosphide, (FeNi)₃P, dissolution in the α or γ phase is diffusion controlled at 975°C. The ternary dissolution paths observed in each of the diffusion couples are unique and the same as those observed in the comparable alloys. The dissolution rate of (FeNi)₃P is controlled by the diffusion rate of P in the α or y phases. The Ni interface compositions in (FeNi)₃P and α or γ and the dissolution path through the ternary are determined by the rate of dissolution and the major Ni ternary diffusion coefficients. It is possible to calculate both the dissolution path and rate for (FeNi)₃P by using the binary dissolution equations in combination with the Fe-Ni-P diagram and the major (ternary) diffusion coefficients. In addition, numerical solutions can be correctly calculated for diffusion controlled dissolution where impingement of overlapping gradients occurs. Formerly Graduate Assistant, Department of Metallurgy and Materials Science, Lehigh University, Bethlehem, Pennsylvania     

Diffusion-coefficient measurements in liquid metallic alloys

The value of the diffusion coefficient in the liquid (D _l ) is generally obtained from the measurement of composition profiles ahead of a quenched planar interface. The experimental results show significant scatter. The main reason for this scatter will be shown to be due to the presence of fluid flow in the liquid. Directional-solidification studies in the Al-Cu system have been carried out to first establish the experimental conditions required for diffusive growth. The composition profiles are then measured to obtain the values of D _l for alloy compositions ranging from 4.0 to 24.0 wt pct Cu. The value of D _l =2.4×10⁻³ mm²/s was obtained along the liquidus line, and this result is significantly smaller than the values reported in the literature, which vary from 3.0 to 5.5 × 10⁻³ mm²/s. It is shown that the scatter in the reported values can be correlated with the diameter of the sample used and, thus, with the fluid flow present in their experiments. Detailed experimental procedures to obtain and verify diffusive-growth conditions are outlined, and appropriate analyses of the data are discussed.     

The effects of silicon on the reaction between solid iron and liquid 55 wt pct Al−Zn baths

The effect of various silicon levels on the reaction between iron panels and Al-Zn-Si liquid baths during hot dipping at 610°C was studied. Five different baths were used: 55Al−0.7Si−Zn, 55Al−1.7Si−Zn, 55Al−3.0Si−Zn, 55Al−5.0Si−Zn, and 55Al−6.88Si−Zn (in wt pct). The phases which formed as a result of this reaction were identified as Fe₂Al5 and FeAl3 (binary Fe−Al phases with less than 2 wt pct Si and Zn in solution),T1, T2, T4, T8, andT _5H (ternary Fe−Al−Si phases), andT _5C (a quaternary Fe−Al−Si−Zn phase). Compositional variations through the reaction zone were determined. The phase sequence in the reaction zone of the panel dipped for 3600 seconds in the 1.7 wt pct Si bath was iron panel/(Fe₂Al₅+T ₁)/FeAl₃/(T _5H+T _5C)/overlay. In the panel dipped for 1800 seconds in the 3.0 wt pct Si bath the reaction zone consisted of iron panel/Fe₂Al₅/(Fe₂Al₅+T ₁)/T ₁/FeAl₃/(FeAl₃+T ₂)/T _5H/overlay. In the panel dipped for 3600 seconds in the 6.88 wt pct Si bath the phase sequence was iron panel/Fe₂Al₅/(Fe₂Al₅+T1)/(T1+FeAl3)/(T1+T2)/T2/T8/T4/overlay. The growth kinetics of the reaction zone were also studied. A minimum growth rate for the reaction zone which formed from a reaction between the iron panel and molten Al−Zn−Si bath was found in the 3.0 wt pct Si bath. The growth kinetics of the reaction layers were found to be diffusion controlled in the 0.7, 1.7, and 6.88 wt pct Si baths, and interface controlled in the 3.0 and 5.0 wt pct Si baths. The presence of the interface between theT2/T5H, Fe2Al5/T ₁, orT ₁/FeAl₃ phases is believed responsible for the interface controlled growth kinetics exhibited in the 3.0 and 5.0 wt pct Si baths.     

The interdiffusivity matrix of Fe2O3-CaO-SiO2 melt at 1693 to 1773 K

The diffusion couple method was used at 1693 to 1773 °K on liquid slags with their average compositon of 20 wt pct Fe₂O −₃−35 wt pct CaO-45 wt pet SiO₂. After diffusion runs for 40 min, the samples have been quenched to glassy state. The samples were sectioned, polished, and analyzed by a X-ray micro analyzer. The diffusivities matrix obtained from the penetration curves can be expressed by the following equations,D ³⁰ ₁₀₋₁₀ = 3.27 exp (−50000―RT)(cm²/s)D ³⁰ ₁₀₋₂₀ = -11.1 exp (−50000―RT)(cm²/s)D ³⁰ ₂₀₋₁₀ = 8.30 exp (−56300―RT)(cm²/s)D ³⁰ ₂₀₋₂₀ = 11.5 exp (−56200―RT)(cm²/s) where 10, 20, and 30 mean Fe₂O₃, CaO and SiO₃, respectively and the activation energies are in Cal per mol. The elements obtained satisfy the restriction derived from the second law of thermodynamics. The diffusion-composition paths obtained are consistent with the Cooper's parallelogram.     

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.     

Reaction diffusion and phase equilibria in the V-N system

The formation of phase bands in in situ diffusion couples of the V-N system was studied by the reaction of vanadium sheet with pure nitrogen within the temperature range 1100 °C to 1700 °C and the nitrogen pressure range 2 to 24 bar. Under these conditions, phase bands of β-V₂N and δ-VN_1−x develop. The morphology of the β-V₂N/α-V(N) interface depends on the saturation state of the α-V(N) core. If the nitrogen content in α-V(N) is high, the interface has a jagged appearance, whereas at low nitrogen contents of the α-V(N) phase, the interface is planar. Electron probe microanalysis (EPMA) was used to measure the diffusion profiles within the couples. The homogeneity regions of the nitride phases were established and the phase diagram accordingly corrected. From the growth rates of the phase bands, the mean composition-independent nitrogen diffusivities in β-V₂N and δ-VN_1−x were derived. These diffusivities follow an Arrhenius equation with activation energies of 2.92 (β-V₂N) and 2.93 eV (δ-VN_1−x ). By using δ-VN_1−x as a starting material and a low nitrogen pressure during annealing, it could be shown that the direction of nitrogen diffusion can be reversed, i.e., β-V₂N is formed on the surface of the couple as a result of out-diffusion of nitrogen.     

Numerical modeling of γ precipitate growth during Fe-Ni martensite decomposition at low temperatures (≤400 °C)

A numerical model was developed to simulate Ni composition profiles developed around γ (FeNi) precipitates growing during martensite (α₂) decomposition in Fe-Ni at low temperatures (300 °C to 400 °C). The model is based on the theory of partial interface reaction control of the precipitate growth process. Experimental Ni composition profiles were measured across γ -α₂ interfaces using high spatial resolution analytical electron microscopy. The simulated Ni composition profiles show good agreement with the experimentally measured profiles, indicating that partial interface reaction control of the γ growth is a reasonable assumption. The diffusion coefficients and the interface mobilities were estimated from the simulations. The activation energy for diffusion in the α₂ matrix obtained from the computer model is 0.7 eV with an error range from 0.58 to 0.98 eV. This value is similar to the activation energy for diffusion obtained from the calculated γ -α₂ interface mobility (0.62 eV with an error range from 0.57 to 0.67 eV). This result is consistent with the observed high dislocation density in the α₂ matrix. Both these values of the activation energy are well below that for lattice diffusion (223C;3 eV). Therefore, it is concluded that the prevailing diffusion mechanisms at these temperatures are short circuit (defect) diffusion in the α₂ matrix and rapid diffusion across the γ -α₂ interface. Formerly Research Assistant, Department of Materials Science and Engineering, Lehigh University     

Reaction diffusion and phase equilibria in the V-N system

The formation of phase bands in in situ diffusion couples of the V-N system was studied by the reaction of vanadium sheet with pure nitrogen within the temperature range 1100 °C to 1700 °C and the nitrogen pressure range 2 to 24 bar. Under these conditions, phase bands of β-V₂N and δ-VN_1−x develop. The morphology of the β-V₂N/α-V(N) interface depends on the saturation state of the α-V(N) core. If the nitrogen content in α-V(N) is high, the interface has a jagged appearance, whereas at low nitrogen contents of the α-V(N) phase, the interface is planar. Electron probe microanalysis (EPMA) was used to measure the diffusion profiles within the couples. The homogeneity regions of the nitride phases were established and the phase diagram accordingly corrected. From the growth rates of the phase bands, the mean composition-independent nitrogen diffusivities in β-V₂N and β-VN_1−x were derived. These diffusivities follow an Arrhenius equation with activation energies of 2.92 (β-V₂N) and 2.93 eV (δ-VN_1−x ). By using δ-VN_1−x as a starting material and a low nitrogen pressure during annealing, it could be shown that the direction of nitrogen diffusion can be reversed, i.e., β-V₂N is formed on the surface of the couple as a result of out-diffusion of nitrogen.     

Diffusion of the tracers Cu67, Ni66, and Zn65 in copper-rich solid solutions in the system Cu-Ni-Zn

Tracer diffusion coefficients were determined for the three isotopes, Zn⁶⁵, Cu⁶⁷, and Ni⁶⁶, in homogeneous Cu-Ni-Zn binary and ternary alloys, to 30 pct Ni and Zn, and pure copper as a function of composition and as a function of temperature, within about 250°C of the solidus surface. Activation energies andD ₀ factors were determined as functions of composition from these measurements. It is found that as the composition plane is traversed in the general direction from high nickel compositions on the copper-nickel binary to high zinc concentrations on the copper-zinc binary,i.e., as nickel is replaced by zinc, the diffusivity of all three tracers increases, and the activation energy for diffusion decreases. The total change in diffusivity across the composition plane is about two orders of magnitude. The three diffusivities are always in the order:D*_Zn >D*_Cu >D*_Ni, with the ratio being 9∶3∶1 at 900°C for all compositions. The three activation energies are usually in the orderQ*_Ni >Q*_Cu >Q*_Zn. These results are shown to be consistent with atom size and electron-to-atom concentrations of the three species in this alloy system. K. J. ANUSAVICE, formerly Graduate Student, Department of Materials Engineering, University of Florida, Gainesville, Fla.     

Solute diffusion in alpha- and gamma-iron

The diffusion rates of chromium, vanadium, and hafnium in α- and γ-Fe have been determined by radiotracer techniques. The results are (in sq cm sec⁻¹): α-Fe γ-Fe ChromiumD = 8.52 exp (−59,900/RT)D = 10.80 exp(−69,700/RT) VanadiumD = 3.92 exp (−57,600/RT)D = 0.25 exp (−63,100/RT) HafniumD = 1.31 exp (−69,300/RT)D = 3600 exp (−97,300/RT) The differences in diffusion rates are discussed in terms of the compressibility of the diffusing atom. Diffusion of chromium in γ-Fe was also measured by a microprobe analysis technique. The result is:D = 4.08 exp (−68,500/RT) Comparison is made between diffusion analysis by tracer techniques and by electron probe microanalysis. Formerly with Department of Metallurgy, University of Manchester, Manchester, England     

Thermodynamic modeling of binary and ternary metallic solutions

A model is proposed for describing heat of mixing behavior in binary and ternary metallic solutions. The binary model, which has the form, ΔH ^M =α₁ X _A ² X _B +α₂ X _A X _B ² −α₃ X _A ² X _B ² , whereX _A andX _B are mole fractions of componentsA andB and α₁, α₂, and α₃ are constants, is applied to the heat of mixing values for 84 solid and liquid systems and the results are compared with the subregular model. The ternary model, which is composed of the sum of the binary equations and a ternary interaction term of the form α _ABC X _A X _B X _C , was applied to the Bi−Cd−Pb, Cd−Pb−Sn, and Cd−Pb−Sb systems. There was excellent agreement both as to the shapes of the isoenthalpy of mixing curves and as to the heat of mixing values in the ternary systems when the model was used to predict the experimental values.     

Activities in the spinel solid solution Fe X Mg1−X Al2O4

Activities in the spinel solid solution Fe _X Mg_1−X Al₂O₄ saturated with α-Al₂O₃ have been measured for the compositional range 0<X<1 between 1100 and 1350 K using a bielectrolyte solid-state galvanic cell, which may be represented as Pt, Fe + Fe _X Mg_1−X Al₂O₄+α-Al₂O₃//(Y₂O₃)ThO₂/(CaO)ZrO₂//Fe + FeAl₂O₄+α-Al₂O₃, Pt Activities of ferrous and magnesium aluminates exhibit small negative deviations from Raoult’s law. The excess free energy of mixing of the solid solution is a symmetric function of composition and is independent of temperature: ΔG ^E=−1990 X(1−X) J/mol. Theoretical analysis of cation distribution in spinel solid solution also suggests mild negative deviations from ideality. The lattice parameter varies linearly with composition in samples quenched from 1300 K. Phase relations in the FeO-MgO-Al₂O₃ system at 1300 K are deduced from the results of this study and auxiliary thermodynamic data from the literature. The calculation demonstrates the influence of intracrystalline ion exchange equilibrium between nonequivalent crystallographic sites in the spinel structure on intercrystalline ion exchange equilibrium between the monoxide and spinel solid solutions (tie-lines). The composition dependence of oxygen partial pressure at 1300 K is evaluated for three-phase equilibria involving the solid solutions Fe + Fe _X Mg_1−X Al₂O₄+α-Al₂O₃ and Fe + Fe _y Mg_1−Y O+Fe _X Mg_1−X Al₂O₄. Dependence of X, denoting the composition of the spinel solid solution, on parameter Y, characterizing the composition of the monoxide solid solution with rock salt structure, in phase fields involving the two solid solutions is elucidated. The tie-lines are slightly skewed toward the MgAl₂O₄ corner.     

Effects of sulfur on the fatigue and fracture resistance of interfaces between γ-Ni(Cr) and α-Al2O3

Dramatic effects of S on the adhesion and fatigue resistance of interfaces between γ-Ni(Cr) and α-Al₂O₃, have been explicitly demonstrated and quantified. This has been achieved by using two bonding conditions: one involving solid-state diffusion (SSDB) and another through a liquid phase (LPB) that forms at temperatures above a eutectic that releases a S-rich liquid. Upon SSDB, there is no significant S excess at the interface, whereas LPB forms a thin interphase with a high local concentration of S. The SSDB materials have interfaces with such high toughness (above 300 J m⁻²) and fatigue resistance that mode I cracks divert into the Al₂O₃, rather than propagate at the interface. Conversely, the LPB materials delaminate at the interface, solely as a result of the residual stresses from thermal expansion misfit, with an interface toughness in the range 2 to 7 J m⁻².     

Application of the diffusion couple to study phase equilibria in the Fe-Cr-S ternary system at 600 °C

The sulfidation of Fe-Cr alloys has a large financial significance for industries that use fossil fuels, such as the electric utility industry. Therefore, the sulfidation of a series of Fe-Cr alloys was studied at 600 °C using a solid-state diffusion couple technique. The diffusion couple technique combined Fe_0.95S powder and FeCr binary alloys together in a configuration that allowed for post-heat-treatment microanalysis using an electron probe microanalyzer (EPMA). The results showed that only two different diffusion couple microstructures formed in samples spanning the entire Fe-Cr binary range. The Fe-rich alloy diffusion couples contained a surface αFeCr layer and an internal sulfide precipitate layer that contained three different sulfide phases. The Cr-rich alloy diffusion couples also possessed an internal precipitate layer, as well as a thick, triplex interfacial scale. The ternary elemental diffusion was described using diffusion paths plotted on the 600 °C isothermal section of the Fe-Cr-S phase diagram. The results also showed that samples with less than 51 wt Pct Cr were more sulfidation resistant. The accuracy of the existing Fe-Cr-S 600 °C isothermal section was assessed, and it was determined that the τ phase field had a larger composition than previously published.     

Precipitation-strengthened austenitic Fe−Mn−Ti alloys

The precipitation of intermetallic compounds in the Fe−20Mn−2Ti and Fe−28Mn−2Ti alloy systems has been investigated over the temperature range 700 to 900°C by hardness measurements, optical and scanning electron microscopy, and X-ray diffraction. In both systems only the equilibrium Laves phase was observed. The precipitate was identified as C14(MgZn₂) type hexagonal Laves phase with a chemical composition close to Fe₂ (Ti, Mn). In an as-annealed sample precipitation occurred in a heterogeneous manner, predominantly along grain boundaries. The effect of a cold deformation between the solution annealing and aging processes was also investigated. In addition to a high density of dislocations, martensitic phases were induced by deformation: a γ→∈ transformation occurred in the Fe−28Mn−2Ti alloy while a γ→α′ transformation was predominant in the Fe−20Mn−2Ti alloy. Subsequent aging was conducted at temperatures above theA _f . A large number of very fine precipitates formed randomly in the matrix after a short aging period. This cold work plus aging treatment resulted in an increase in yield strength. The enhancement of mechanical properties is due to the randomly distributed precipitates combined with the high defect density and fine substructure.     

Diffusion of managenese and silicon in liquid iron over the whole range of composition

The measurement of the diffusivities of manganese and silicon in molten binary ferroalloys over the whole range of composition was undertaken to clarify existing but conflicting data at lower concentrations, to present new data at higher concentrations and to indirectly confirm the behavior of both systems observed in thermodynamic studies. The experiments were carried out under argon atmosphere in a Tammann furnace. The diffusion couples were held in 5 mm ID alumina tubes (98 pct Al₂O₃). Electron probe microanalysis of the samples led to a diffusion-penetration curve for the system under consideration. Results obtained over the whole range of composition showed a slight negative deviation for the Fe−Mn system and a very large positive deviation for the Fe−Si system. At lower concentrations (0 to 4 pct Mn), the temperature dependence of managanese diffusivity for the Fe−Mn binary alloy in the temperature range 1550° to 1700°C is as follows:D _Fe−Mn=1.8×10⁻³ exp (−13,000/RT) cm²/sec The concentration dependence of manganese diffusivity for the same system at 1600°C may be expressed asD _Fe−Mn={5.48−0.0137 (%Mn)+0.000276 (%Mn)²}×10⁻⁵ cm²/sec The temperature dependence of silicon diffusivity for the Fe−Si binary system in the temperature range 1550° to 1725°C at various concentrations is as follows:D _Fe−Si=2.8×10⁻³ exp (−11,900/RT) cm²/sec at 20 pct SiD _Fe−Si=2.1×10⁻³ exp (−13,200/RT) cm²/sec at 12.5 pct SiD _Fe−Si=5.1×10⁻⁴ exp (−9,150/RT) cm²/sec at 2.2 pct Si FELIPE P. CALDERON, formerly Graduate Student. University of Tokyo, Tokyo, Japan. This paper is based on a portion of a thesis submitted by FELIPE P. CALDERON in partial fulfillment of the requirements for the degree of Doctor of Engineering at University of Tokyo.     

Interface stability during displacement reactions between Cu<Subscript>2</Subscript>O and Co<Subscript>1−<Emphasis Type="Italic">X</Emphasis></Subscript>Fe<Subscript><Emphasis Type="Italic">X</Emphasis></Subscript> alloys at 1223 K

The stability of the reactive interface during the solid-state displacement reaction, Cu₂O+Co_1−X Fe _X =2Cu+(Co_1−X Fe _X )O, is studied as a function of Co-Fe alloy composition at 1223 K. For X≤0.03, the reaction zone has a layered structure, and the cation diffusion in (Co, Fe)O is the rate-limiting step. The interface is unstable in the early stages of the reaction; the instability decreases with time as the oxide thickness increases, and the interface becomes planar at long times. The time required for the attainment of interface planarity increases with the value of X. The reaction kinetics are consistent with the available cation-diffusion data in (Co, Fe)O. For X≥0.045, the product zone is a composite of Cu+(Co, Fe)O, and the rate is limited by the oxygen transport in copper. The transition to interface instability occurs when the oxide can support a cation flux that exceeds the maximum possible oxygen flux in copper. During the reaction, composition gradients develop in (Co, Fe)O because of higher diffusion rates for iron than for cobalt.