Kinetics of phase evolution of Zn-Fe intermetallics

The intermetallic phases, Γ (Fe₃3Zn₁₀), Γ₁, (Fe₅Zn₂₁), δ (FeZn₇, and ζ (FeZn₁₃), are mechanically alloyed through ball milling of pure elemental Fe and Zn powders under a controlled atmosphere of argon gas. The state of the as-ball-milled materials was crystalline, except for the Γ phase, which was amorphous. Phase-evolution kinetics through differential scanning calorimeter (DSC) measurements of the as-ball-milled powders show three characteristic transition temperatures for the Γ₁, and ζ phases, two for the Γ phase, and only one for the δ phase. The activation energies for the evolution of the milled powders to their equilibrium crystalline phases are 170 ± 1, 251 ±2, 176± 1, and 737 ±4 kJ/mol for the Γ, Γ₁, δ, and ζ phases, respectively. These values show that the mechanisms for the metastable-to-stable phase transition in these intermetallics are different, whereas diffusion over short distances may be part of the entire process in all cases.     

Thermal transformations in mechanically alloyed Fe-Zn-Si materials

The ball milling of elemental powders corresponding to Γ (Fe₃Zn₁₀)+0.12 wt pct Si; Γ₁ (Fe₅Zn₂₁) + 0.12 wt pct Si; δ (FeZn₇)+0.12 wt pct Si; and ζ (FeZn₁₃)+0.12 wt pct Si composition ratios yields crystalline, mechanically alloyed phases. Differential scanning calorimetry (DSC) measurements of these materials show that they evolve differently, with well-defined characteristic stages. The activation energies for processes corresponding to these stages, based on kinetic analyses, are determined and correlated to microstructural evolvements. The processes occurring during the first stage below 250 °C, for all of the materials studied using X-ray diffraction (XRD) analysis, are associated with release of strain, recovery, and limited atomic diffusion. The activation energies for recovery processes are 120 kJ/mole for the Γ+0.12 wt pct Si, 131 kJ/mole for δ+0.12 wt pct Si, and 96 kJ/mole for ζ+0.12 wt pct Si alloys. At higher temperatures, recrystallization and other structural transformations occur with activation energies of 130 and 278 kJ/mole for Γ+0.12 wt % Si; of 161 kJ/mole for Γ₁+0.12 wt pct Si; of 167 and 244 kJ/mole for δ+0.12 wt pct Si; and of 641 kJ/mole for the ζ+0.12 wt pct Si. In addition, a eutectic reaction at 420 °C±3 °C, corresponding to the Zn-Si system, and a melting of Zn in Fe-Zn systems are observed for the ζ+0.12 wt pct Si material. The relation of FeSi formation in the Sandelin process is discussed.     

Kinetics and phase transformation evaluation of Fe-Zn-Al mechanically alloyed phases

Fixed composition ratios of Fe and Zn corresponding to γ-(Fe₃Zn₁₁₀), Γ₁-(Fe₅Zn₂₁), δ-(FeZn₇), and ζ-(FeZn₁₃) with the addition of 5 pct Al (wt) were ball milled in an argon gas atmosphere to form homogenous alloys. Nonisothermal kinetic analyses of the mechanically alloyed materials, based on differential scanning calorimetry (DSC) measurements, revealed two diffusion-controlled processes during the evolution of the δ+5 pct Al and ζ+5 pct Al compositions with activation energies of 227±2 and 159±1 kJ/mole, respectively. Other endothermic and exothermic reactions detected for these compositions are consistent with the Fe-Zn-Al equilibrium phase systems with respect to the formation of the Fe₃Al, Fe₂Al₅, and δ-FeZn₇ phases Based on the evidence of FeAl₂ formation at 440 °C for the ζ+5 pct Al composition from X-ray diffraction (XRD) and DSC measurements, the revision/re-evaluation of the Fe-Zn-Al equilibrium phase diagrams is proposed. The Γ+5 pct Al and Γ₁+5 pct Al compositions evolved similarly through the same fields, except at 400 °C, where the former consisted of α-Fe + Γ + δ, while the later was without the Γ phase.     

The effect of steel chemistry on the formation of Fe-Zn intermetallic compounds of galvanneal-coated steel sheets

The effects of steel chemistry on the formation of Fe-Zn intermetallic compounds in the galvanneal coatings have been investigated by examining the microstructure of galvanneal coat-ings on extra-low-carbon (ELC) steel, interstitial-free (IF) steel, and interstitial-free rephos-phorized (IFP) steel. The layer structure of the coatings was revealed by chemical etching. Phases present in each layer were then identified using electron diffraction in transmission elec-tron microscopy (TEM). A two-layer structure, one consisting of the δ phase with a small fraction of the ζ, phase dispersed on the surface and Γ phases and another consisting of the δ and Γ₁ phases, was observed in the ELC sample and the IFP sample, respectively. A three-layer structure consisting of the δ, Γ₁ + δ, and Γ phases was observed in the IF sample. The presence of C in the steel substrate retarded the alloying between Fe and Zn; while P in the steel favored the formation of the Γ₁, phase over the Γ phase by its surface segregation in the steel substrate. The orientation relationship between coating and substrate was also studied by electron diffraction. Three α-Fe/Γ orientation relationships were frequently observed.     

Microstructural study of galvanized coatings formed in pure as well as commercial grade zinc baths

This investigation is an effort to have a better understanding of the growth kinetics and morphology of the coating formed during the galvanizing process in pure as well as commercial grade zinc baths. The protective coating that is formed during hot dip galvanizing, normally between 450 and 480°C, consists of a series of Fe-Zn intermetallic layers, which have been identified as gamma (Γ), delta (δ), zeta (ξ) and an outer eta (η) layer, highly rich in zinc. There is apparently no delay in the formation of ξ or δ phases in both pure as well as the commercial grade zinc baths. The gamma (Γ) phase is formed after an incubation time of about 30 s at a bath temperature of 470°C in the pure zinc bath. Its formation is further delayed in the commercial grade zinc bath. The last morphological feature is the formation of a second ξ layer at the ξ/δ interface in the pure zinc bath. In the commercial grade zinc bath two different morphologies of ξ phase are seen starting from the lowest dipping time, and also the overall coating is considerably thicker due to formation of several iron-zinc intermetallics which degrade its ductility and outward appearance. Commercial grade zinc also enhances the dross formation in the bath and deteriorates the quality of the coating. Presence of transverse cracks as well as entrapment of dross particles in the coating is attributed to the less compact coating that is formed in the commercial grade zinc bath.     

Microstructure and properties of duplex δ-Al3(Ti, V)/β-(Ti, V) alloys

The microstructures of multiphase intermetallic alloys with compositions Al₇₀Ti₁₀V₂₀ and Al₆₂Ti₁₀V₂₈ based on the trialuminide Al₃Ti have been characterized, following chill casting and postsolidification heat treatment, using a combination of scanning electron microscopy and transmission electron microscopy (TEM). Evidence of a eutectic reaction of the form L → δ-Al₃(Ti, V)+ζ-Al₈V₅, not previously reported in the Al-Ti-V system, has been observed in both alloys solidified at sufficient levels of undercooling. The ζ phase is replaced by metallic β-(Ti, V) phase during subsequent heat treatment in the range 1073 to 1273 K, and differential thermal analysis (DTA) of samples prean-nealed at 1173 K revealed an endothermic peak at ∼1560 K, consistent with equilibrium eutectic melting of the form (δ+β) → L. Although the chill-cast alloys retained metastable intermediate high-temperature phases, duplex metallic-intermetallic microstructures, containing uniform fine-scale distributions of metallic β-(Ti, V) solid solution in a δ-Al₃(Ti, V) intermetallic matrix, have been produced in both alloys during isothermal heat treatments at temperatures in the range 1073 to 1273 K. For both alloys, the bulk Vickers hardness of such microstructures remained in excess of that of binary Al₃Ti, while in the Al₆₂Ti₁₀V₂₈ alloy, where the increased volume fraction of β phase took the form of a near-continuous network within δ matrix, there was evidence arising from indentation tests of a substantial improvement in the cracking resistance compared to both chill-cast ternary alloy and binary Al₃Ti.     

Distribution of aluminum in hot-dip galvanized coatings

Hot-dip galvanized panels of low-carbon (LC) and interstitial-free (IF) steels were produced in a laboratory simulator with an average coating mass of 60 g/m². Three pot aluminum levels were used, viz., 0.10 pct (by wt), 0.15 pct, and 0.18 pct. Metallography, scanning electron microscopy (SEM), and transmission electron microscopy (TEM) were used to characterize coating and base steel microstructures. Wet chemical analysis and scanning transmission electron microscopy (STEM) were employed for compositional analyses. The aluminum content of the melt was found to be the predominant factor influencing the distribution of Al in the coating. At 0.18 pct melt aluminum, Al is partitioned between the aluminide inhibition layer at the coating-steel interface (∼80 pct) and the zinc overlay (∼20 pct). At 0.15 pct, it is partitioned among the aluminide layer (∼75 pct to 80 pct), zinc-iron (FeZn₁₃, ζ) intermetallic layer (∼5 pct to 15 pct), and the coating overlay (∼10 pct). At 0.10 pct, the aluminum is divided almost equally between the overlay and the zinc-iron intermetallics. At the two lower aluminum levels is the distribution marginally influenced by the steel grade. The ζ was found to not preferentially nucleate at the ferrite grain boundaries. When both the aluminide and ζ occurred at the coating-steel interface, the ζ particles appeared near discontinuities and thinner regions in the aluminide layer. The coating, relative to the melt, is enriched in aluminum because of its concentration in the aluminide and in the zinc-iron intermetallics. This enrichment increases with melt aluminum through an increase in the aluminum content of the aluminide layer and not of its thickness. In addition, a few tens-of-nanometers-thick layer enriched in aluminum, oxygen, and iron is observed on the outer surface of all coatings. The aluminum content in this layer also increases with an increase in the melt aluminum, but it contributes negligibly to the coatings’s content because of its extreme thinness.     

Iron intermetallic phases in the Al corner of the Al-Si-Fe system

The iron intermetallics observed in six dilute Al-Si-Fe alloys were studied using thermal analysis, optical microscopy, and image, scanning electron microscopy/energy dispersive X-ray, and electron probe microanalysis/wavelength dispersive spectroscopy (EPMA/WDS) analyses. The alloys were solidified in two different molds, a preheated graphite mold (600 °C) and a cylindrical metallic mold (at room temperature), to obtain slow (∼0.2 °C/s) and rapid (∼15 °C/s) cooling rates. The results show that the volume fraction of iron intermetallics obtained increases with the increase in the amount of Fe and Si added, as well as with the decrease in cooling rate. The low cooling rate produces larger-sized intermetallics, whereas the high cooling rate results in a higher density of intermetallics. Iron addition alone is more effective than either Si or Fe+Si additions in producing intermetallics. The alloy composition and cooling rate control the stability of the intermetallic phases: binary Al-Fe phases predominate at low cooling rates and a high Fe:Si ratio; the β-Al₅FeSi phase is dominant at a high Si content and low cooling rate; the α-iron intermetallics (e.g., α-Al₈Fe₂Si) exist between these two; while Si-rich ternary phases such as the δ-iron Al₄FeSi₂ intermetallic are stabilized at high cooling rates and Si contents of 0.9 wt pct and higher. Calculations of the solidification paths representing segregations of Fe and Si to the liquid using the Scheil equation did not conform to the actual solidification paths, due to the fact that solid diffusion is not taken into account in the equation. The theoretical models of Brody and Flemings^[44] and Clyne and Kurz^[45] also fail to explain the observed departure from the Scheil behavior, because these models give less weight to the effect of solid back-diffusion. An adjusted 500 °C metastable isothermal section of the Al-Si-Fe phase diagram has been proposed (in place of the equilibrium one), which correctly predicts the intermetallic phases that occur in this part of the system at low cooling rates (∼0.2 °C/s).     

Influence of Aluminum on the Formation Behavior of Zn-Al-Fe Intermetallic Particles in a Zinc Bath

The shape, size, and composition of dross particles as a function of aluminum content at a fixed temperature were investigated for aluminum added to the premelted Zn-Fe melt simulating the hot-dip galvanizing bath by a sampling methodology. In the early stage, less than 30 minutes after Al addition, local supersaturation and depletion of the aluminum concentration occurred simultaneously in the bath, resulting in the nucleation and growth of both Fe₂Al₅Zn _x and FeZn₁₃. However, the aluminum was homogenized continuously as the reaction proceeded, and fine and stable FeZn₁₀Al _x formed after 30 minutes. An Al-depleted zone (ADZ) mechanism was newly proposed for the “η→η+ζ→δ” phase transformations. The ζ phase bottom dross partly survived for a relatively long period, i.e., 2 hours in this work, whereas the η phase disappeared after 30 minutes. In the early stage of dross formation, both Al-free large particles as well as high-Al tiny particles were formed. The dross particle size decreased slightly with increased reaction time before reaching a plateau. The opposite tendency was observed when the Al content was 0.130 mass pct; with a relatively high Al content, the nucleation of tiny η phase dross was significantly enhanced because of the high degree of supersaturation. This unstable η phase dissolved continuously and underwent simple transformation to the stable δ phase. The relationship between nucleation potential and supersaturation ratio of species is discussed based on the thermodynamics of classical nucleation theory.     

Iron intermetallic phases in the Al corner of the Al-Si-Fe system

The iron intermetallics observed in six dilute Al-Si-Fe alloys were studied using thermal analysis, optical microscopy, and image, scanning electron microscopy/energy dispersive X-ray, and electron probe microanalysis/wavelength dispersive spectroscopy (EPMA/WDS) analyses. The alloys were solidified in two different molds, a preheated graphite mold (600°C) and a cylindrical metallic mold (at room temperature), to obtain slow (}0.2 °C/s) and rapid (}15 °C/s) cooling rates. The results show that the volume fraction of iron intermetallics obtained increases with the increase in the amount of Fe and Si added, as well as with the decrease in cooling rate. The low cooling rate produces larger-sized intermetallics, whereas the high cooling rate results in a higher density of intermetallics. Iron addition alone is more effective than either Si or Fe+Si additions in producing intermetallics. The alloy composition and cooling rate control the stability of the intermetallic phases: binary Al-Fe phases predominate at low cooling rates and a high Fe:Si ratio; the β-Al₅FeSi phase is dominant at a high Si content and low cooling rate; the α-iron intermetallics (e.g., α-Al₈Fe₂Si) exist between these two; while Si-rich ternary phases such as the δ-iron Al₄FeSi₂ intermetallic are stabilized at high cooling rates and Si contents of 0.9 wt pct and higher. Calculations of the solidification paths representing segregations of Fe and Si to the liquid using the Scheil equation did not conform to the actual solidification paths, due to the fact that solid diffusion is not taken into account in the equation. The theoretical models of Brody and Flemings^[44] and Clyne and Kurz^[45] also fail to explain the observed departure from the Scheil behavior, because these models give less weight to the effect of solid back-diffusion. An adjusted 500°C metastable isothermal section of the Al-Si-Fe phase diagram has been proposed (in place of the equilibrium one), which correctly predicts the intermetallic phases that occur in this part of the system at low cooling rates (}0.2 °C/s).     

The C.L.m-δ equation of superplasticity

In the fundamental equation of superplastic flow, σ = kέ^m, έ is the strain rate, σ is the flow stress,m is the strain-rate sensitivity index of flow stress, andk is a material constant. Bothm andk vary with increase in strain (δ). Therefore, we can have m0 (≠0), m_I (m_I1,m,_I2,m_I3,....), m_F; k₀ (≠0), k_I (k_I1,k_I2,k_I3,.... ), k_F values corresponding to the initial strain δ₀ (= 0.00 pct), the strain at any instant during stretching, δ_I (δ_I1, δ_I2, 8δ_I3,....), and the total elongation, δ_F, at fracture, respectively. The curves expressing the dependence ofm andk on δ can be called them-k- δ orm-C- δ or simply them- δ curves. All these curves can be classified into types and can be expressed by the C. L.m- δ equation derived by the present author: δ(pct) = [Cέ^{(m- m0} - 1] × 100 whereC = k₀ + dk₀/k₀,m = m₀ andC = k₀/k₀ = 1, when δ = δ₀;m = m_I, andC = C_I = k_I/k₀ when δ= δ_I;m = m_F andC = C_F = k_F/k₀, when δ = δ_F. By means of this equation, not only the total elongation at fracture but also the strains at any instant during stretching of a material can be predicted, if the correspondingm values are known.     

A study of the thermodynamics and phase relationships of the Fe−Co−S−O quaternary system

The oxygen potentials of several three- and two-phase equilibria in the Fe−Co−S−O quaternary system were measured atP _SO2=1, 0.1, and 0.01 atm over wide temperature ranges. The measurements were carried out using a solid oxide electrolyte emf technique. The equilibria measured are sp+ε+δ, s+ε+ξ, sp+ξ+η, and sp+δ. The symbols sp, ε, δ, ξ, and η denote the spinel, monoxide, monosulfide, metal sulfate, and Fe₂O₃ phases, respectively. Compositions for several of the equilibrated phases were measured using electron probe microanalysis. The present results and literature data for the constituent ternary systems were used to obtain thermochemical solution parameters for the sp, ε, δ and ξ solid solution phases. The calculated potential-composition stability diagrams for SO₂ pressures of 1, 0.1, and 0.01 atm at 973, 1023, and 1073 K, respectively, are in good agreement with the experimental results. OMRAN A. MUSBAH, formerly Research Associate at the University of Wisconsin-Madison     

A pressure-composition-temperature study of Zr-Nb-H system

The solubility of hydrogen was determined in the (Zr + 5 wt pct Nb)-H₂, (Zr + 10 wt pct Nb)-H₂, and (Zr + 20 wt pct Nb)-H₂ systems as a function of composition, temperature (700° to 950°C) and hydrogen equilibrium pressure (0.5 to 760 mm Hg). The position of boundariesβ - (β + δ) and(β + δ)-δ were determined in each of the above three systems. Niobium significantly reduces the solubility of hydrogen in theβ andδ phases and increases the equilibrium hydrogen pressure for any fixed concentration. The equilibrium pressure-temperature relations in the two phase region (β + δ) were derived and the heat of formation ofδ-hydride from saturatedβ-Zr, ΔH _β → δ, were determined. The value of ΔH _β → δ increases up to about 5 wt pct Nb after which the effect of niobium seems to be insignificant. The maximum hydrogen pick-up of zirconium at room temperature decreases with increasing niobium content of the alloy.     

Influence of interface microstructure on the strength of the transition joint between Ti-6Al-4V and stainless steel

Solid-state diffusion bonding of Ti-6Al-4V and type 304 SS was investigated in the temperature range of 750 °C to 950 °C, under a uniaxial load for 5.4 ks in vacuum. The diffusion bonds were characterized using light and scanning electron microscopy. The scanning electron microscopic images in backscattered mode show the existence of different reaction layers in the diffusion zone. The composition of these layers was determined by energy-dispersive X-ray spectroscopy (EDS) to contain the α-Fe, χ, λ, FeTi, β-Ti, and Fe₂Ti₄O phases. The presence of these intermetallics was confirmed by X-ray diffraction. The bond strength was evaluated, and the maximum tensile strength of ∼342 MPa and the maximum shear strength of ∼237 MPa were obtained for the diffusion couple processed at 800 °C due to the finer width of the brittle intermetallic layers. With a rise in joining temperature, the bond strength drops owing to an increase in the width of the reaction layers. The activation energy and growth constant were calculated in the temperature range of 750 °C to 950 °C for the reaction products. The χ phase showed the fastest growth rate. A fracture-surface observation in a scanning electron microscope (SEM) using EDS demonstrates that failure takes place mainly through the β-Ti phase for the diffusion couples processed in the aforementioned temperature range.     

The role of oxides in the formation of primary iron intermetallics in an Al-11.6Si-0.37Mg alloy

Recent research suggest that the iron-rich intermetallic phases, such as α-Fe Al₁₅(Fe,Mn)₃Si₂ and β-Fe Al₅FeSi, nucleate on oxide films entrained in aluminum casting alloys. This is evidenced by the presence of crack-like defects within these iron-rich intermetallics. In an attempt to verify the role of oxides in nucleating iron-rich intermetallics, experiments have been conducted under conditions where in-situ entrained oxide films and deliberately added oxide particles were present. Iron-rich intermetallics are observed to be associated with the oxides in the final microstructure, and crack-like defects are often observed in the β-Fe plates. The physical association of the Fe-rich intermetallic phases with these solid oxides, either formed in situ or added, is in accordance with the mechanism suggesting that iron-rich intermetallics nucleate upon the wetted sides of double oxide films. This article is based on a presentation made in the John Campbell Symposium on Shape Casting, held during the TMS Annual Meeting, February 13–17, 2005, in San Francisco, CA.     

Examination of solidification pathways and the liquidus surface in the Nb-Ti-Al system

The solidification pathways, subsequent solid-state transformations, and the liquidus surface in the Nb-Ti-Al system have been examined as part of a larger investigation of phase equilibria in Nb-Ti-Al intermetallic alloys. Fifteen alloys ranging in composition from 15 to 40 at. pct Al, with Nb to Ti ratios of 4:1, 2:1, 1.5:1, 1:1, and 1:1.5, were prepared by arc melting and the as-cast microstructures were characterized by optical microscopy (OM), microhardness, X-ray diffraction (XRD), differential thermal analysis (DTA), backscattered electron imaging (BSEI), electron probe microanalysis (EPMA), and transmission electron microscopy (TEM). The results indicate that the range of primary β solidification is much wider than that indicated in previously reported liquidus surfaces, both experimental and calculated. Differential thermal analysis has identified the existence of a β to σ+γ transformation in three alloys where it was previously thought not to exist; confirmation was provided by high-temperature vacuum heat treatments in the single-phase β region followed by rapid quenching. The location of the boundary between the β, σ, and δ primary solidification fields has been redefined. A massive β → δ transformation, which was observed in the cast microstructure of a Nb-25Ti-25Al alloy, was repeatable through cooling following homogenization. A β → δ+σ eutectoid-like transformation in the 25 at. pct Al alloys, was detected by DTA and evaluated through microstructural analysis of heat-treated samples. Trends in the β phase with variations in composition were established for both lattice parameters and microhardness. As a result of this wider extent of the primary β solidification field, a greater possibility exists for microstructural control through thermal processing for alloys consisting of either σ+γ, β+σ, or β+δ phases. An erratum to this article is available at .     

Enhanced ductility in Al-Si-Cu-Mg foundry alloys with high Si content

It has recently been suggested that the β-Al₅FeSi and ϑ-Al₂Cu intermetallic particles are refined and dispersed in the presence of high silicon, thereby improving the ductility of Al-Si-Cu-Mg alloys. However, limited metallographic evidence was presented to support these claims. Therefore, a study of the effect of Si content in the range of 4.5 to 9 pct on the morphology and distribution of Fe-rich and Cu-rich intermetallic phases has now been conducted. It is shown that Si, indeed, exerts a refining effect on the iron-containing particles (α and β) and disperses clusters of intermetallics (including the Cu-rich particles). In alloys with low Si content, the Fe- and Cu-rich particles form long and closely intertwined clusters. Microcracks originating from cracked intermetallic particles extend and propagate along the clusters with little plasticity, resulting in the low ductility of the alloys. At a high Si content, the intermetallic phases appear more dispersed and the clusters of particles are small and isolated from each other. Microcracks resulting from the cracked intermetallics are short and are isolated, as well, thereby increasing the ductility of the alloys. The mechanisms by which the refinement and dispersion of intermetallic phases occur are discussed. This article is based on a presentation made in the John Campbell Symposium on Shape Casting, held during the TMS Annual Meeting, February 13–17, 2005, in San Francisco, CA.