On the influence of in-situ reactions on grain size during reactive atomization and deposition

A fundamental study of the factors that govern grain size of 5083 Al processed via reactive atomization and deposition (RAD) is reported. Microstructural observation shows that the average grain size in RAD 5083 Al is slightly smaller than that in the material processed via N₂ spray deposition (SDN). A numerical approach, together with measurements of the temperature histories inside the deposited materials, is implemented to analyze the influence of in-situ reactions during RAD process on the evolution of grain size. The numerical results show that RAD 5083 Al possesses a slightly higher density of nuclei relative to that present in SDN 5083 Al on a per unit volume of deposited material basis at the beginning of the slow solidification of remaining liquid phase. Furthermore, the RAD material exhibits a slightly lower coarsening extent during the slow solidification. Grain growth is negligible during the solid-phase cooling. Accordingly, the calculated grain size in RAD 5083 Al is slightly smaller than that in SDN 5083 Al, consistent with the observed results.     

Numerical modeling of solidification and subsequent transformation of Fe-Cr-Ni alloys

A computational method for the analysis of phase transformation involving solidification was developed with the assumption of thermodynamic equilibria at interfaces. The region of interest was divided into finite segments, and solute diffusion across the segments was computed by the use of the direct finite difference method (FDM). Simultaneously, thermodynamic equilibrium at each interface was updated at every step of the diffusion analysis to determine the location of the interfaces. The temperature decrease and the increment of fraction solid were calculated based on thermal balance, including a heat extraction condition. Solid state transformation from δ to γ phase within each FDM segment was modeled by the use of a Clyne-Kurz (C-K) type analysis with assumptions of complete mixing of solutes in theδ phase and limited back diffusion in theγ phase. The calculation results were compared with welding solidification experiments in the iron-chromium-nickel ternary system. Good agreement was obtained with respect to solute distribution and residual fraction ofδ phase over different compositions and solidification modes of the alloys used.     

Processing, microstructure, and mechanical properties of cast In-Situ Al(Mg, Ti)-Al2O3(TiO2) composite

In-situ particle-reinforced aluminum alloy-based cast composites have been synthesized by solidification of the slurry obtained by dispersion of externally added titanium dioxide (TiO₂) particles in molten aluminum at different processing temperatures. Alumina particles (Al₂O₃) form in situ through chemical reaction of TiO₂ particles with molten aluminum. Simultaneously, the chemical reaction also releases titanium, which dissolves into molten aluminum and results in the formation of intermetallic phase Ti(Al_1−x ,Fe _x )₃ during solidification. Increasing the processing temperature increases (1) the amount of elongated as well as blocky intermetallic phase Ti(Al_1−x ,Fe _x )₃, (2) the proportion of alumina particles in the reinforcing oxides, and (3) the porosity content in the resulting cast in-situ composite. The difference in particle content and porosity between the top and the bottom of the cast ingot increases with increasing processing temperature. The hardness of the cast in-situ composite is significantly more than that of the matrix alloy due to the presence of reinforcing particles, but the hardness is greatly impaired by the presence of porosity at the top of the cast ingot. The percent elongation of the cast in-situ composite decreases with increasing processing temperature possibly due to increasing porosity as well as an increasing amount of elongated intermetallic phase, which affects the percent elongation of the matrix alloy. The tensile and yield stresses of the cast in-situ composite decreases with increasing processing temperature again due to increasing porosity, which affects the ultimate tensile stress more than the yield stress. In the cast in-situ composite containing 3.31 ± 0.77 vol pct of porosity, the Brinell hardness is about 6 times its yield stress. The estimated yield stress of the cast in-situ composite at zero porosity as given by the linear least-squares fit appears to increase with particle content at a significantly higher rate than that predicted by the shear-lag model.     

Application of electromagnetic separation of phases in alloy melt to produce <Emphasis Type="Italic">In-situ</Emphasis> surface and functionally gradient composites

The principle of electromagnetic separation of phases (primary phase) in alloy melt is that the electromagnetic force scarcely acts on the primary phases due to its low electric conductivity as compared to the melt. As a result, a repulsive force acts on the primary iron-rich phases to push them to move in the direction opposite to that of the electromagnetic force. The in-situ surface composite and the functionally gradient composite reinforced by primary Si are produced when the hypereutectic Al-Si alloy solidifies under electromagnetic force induced by static magnetic field and DC current. Similarly, the Al-Si-1.20 pct Fe-1.60 pct Mn alloy in-situ surface composite reinforced by primary iron-rich phase is produced. Based on this, a new method for production of in-situ multigradient composite with several layers, by electromagnetic separation of phases and directional solidification technique, is proposed.     

Texture Evolution and Phase Transformation in Titanium Investigated by <Emphasis Type="Italic">In-Situ</Emphasis> Neutron Diffraction

Time-of-flight neutron diffraction was used to study in-situ texture evolution and the α → β phase transformation in cold-drawn titanium upon continuous heating. The texture changes in the α phase at elevated temperatures upon recrystallization are presented. For the first time, a transient β texture was observed during the α → β transformation, as indicated by the initial rise and the final drop of the {110} _β reflection intensity. This unusual observation is explained in terms of competitive growth between inter- and intragranular β allotriomorphs.     

Optimization of the johnson-mehl-avarami equation parameters for <Emphasis Type="Italic">α</Emphasis>-ferrite to <Emphasis Type="Italic">γ</Emphasis>-austenite transformation in steel welds using a genetic algorithm

A nonisothermal Johnson-Mehl-Avarami (JMA) equation with optimized JMA parameters is proposed to represent the kinetics of transformation of α-ferrite to γ-austenite during heating of 1005 steel. The procedure used to estimate the JMA parameters involved a combination of numerical heat-transfer and fluid-flow calculations, the JMA equation for nucleation and growth for nonisothermal systems, and a genetic algorithm (GA) based optimization tool that used a limited volume of experimental kinetic data. The experimental data used in the calculations consisted of phase fraction of γ-austenite measured at several different monitoring locations in the heat-affected zone (HAZ) of a gas tungsten arc (GTA) weld in 1005 steel. These data were obtained by an in-situ spatially resolved X-ray diffraction (SRXRD) technique using synchrotron radiation during welding. The thermal cycles necessary for the calculations were determined for each monitoring location from a well-tested three-dimensional heat-transfer and fluid-flow model. A parent centric recombination (PCX) based generalized generation gap (G3) GA was used to obtain the optimized values of the JMA parameters, i.e., the activation energy, pre-exponential factor, and exponent in the nonisothermal JMA equation. The GA based determination of all three JMA equation parameters resulted in better agreement between the calculated and the experimentally determined austenite phase fractions than was previously achieved.     

Simulation of time-dependent pool shape during laser spot welding: Transient effects

The shape and depth of the area molten during a welding process is of immense technical importance. This study investigates how the melt pool shape during laser welding is influenced by Marangoni convection and tries to establish general qualitative rules of melt pool dynamics. A parameter study shows how different welding powers lead to extremely different pool shapes. Special attention is paid to transient effects that occur during the melting process as well as after switching off the laser source. It is shown that the final pool shape can depend strongly on the welding duration. The authors use an axisymmetric two-dimensional (2-D) control-volume-method (CVM) code based on the volume-averaged two-phase model of alloy solidification by Ni and Beckermann^[1] and the SIMPLER algorithm by Patankar.^[2] They calculate the transient distribution of temperatures, phase fractions, flow velocities, pressures, and concentrations of alloying elements in the melt and two solid phases (peritectic solidification) for a stationary laser welding process. Marangoni flow is described using a semiempirical model for the temperature-dependent surface tension gradient. The software was parallelized using the shared memory standard OpenMP.     

Microstructural development during solidification of stainless steel alloys

The microstructures that develop during the solidification of stainless steel alloys are related to the solidification conditions and the specific alloy composition. The solidification conditions are determined by the processing method,i.e., casting, welding, or rapid solidification, and by parametric variations within each of these techniques. One variable that has been used to characterize the effects of different processing conditions is the cooling rate. This factor and the chemical composition of the alloy both influence (1) the primary mode of solidification, (2) solute redistribution and second-phase formation during solidification, and (3) the nucleation and growth behavior of the ferrite-to-austenite phase transformation during cooling. Consequently, the residual ferrite content and the microstructural morphology depend on the cooling rate and are governed by the solidification process. This paper investigates the influence of cooling rate on the microstructure of stainless steel alloys and describes the conditions that lead to the many microstructural morphologies that develop during solidification. Experiments were performed on a series of seven high-purity Fe-Ni-Cr alloys that spanned the line of twofold saturation along the 59 wt pct Fe isopleth of the ternary alloy system. High-speed electron-beam surface-glazing was used to melt and resolidify these alloys at scan speeds up to 5 m/s. The resulting cooling rates were shown to vary from 7°C/s to 7.5×10⁶°C/s, and the resolidified melts were analyzed by optical metallographic methods. Five primary modes of solidification and 12 microstructural morphologies were characterized in the resolidified alloys, and these features appear to be a complete “set” of the possible microstructures for 300-series stainless steel alloys. The results of this study were used to create electron-beam scan speedvs composition diagrams, which can be used to predict the primary mode of solidification and the microstructural morphology for different processing conditions. Furthermore, changes in the primary solidification mode were observed in alloys that lie on the chromium-rich side of the line of twofold saturation when they are cooled at high rates. These changes were explained by the presence of metastable austenite, which grows epitaxially and can dominate the solidification microstructure throughout the resolidified zone at high cooling rates. J. W. ELMER, formerly Graduate Student at the Massachusetts Institute of Technology     

The effect of quenching on the solidification structure and transformation behavior of stainless steel welds

Weld solidification structure of three different types of stainless steel,i.e., 310 austenitic, 309 and 304 semiaustenitic, and 430 ferritic, was investigated. Welds of each material were made without any quenching, with water quenching, and with liquid-tin quenching during welding. The weld micro-structure obtained was explained with the help of the pseudobinary phase diagrams for Fe-Cr-Ni and Fe-Cr-C systems. It was found that, due to the postsolidification 5 → γ phase transformation in 309 and 304 stainless steels and the rapid homogenization of microsegregation in 430 stainless steel, their weld solidification structure could not be observed unless quenched from the solidification range with liquid tin. Moreover, the formation of acicular austenite, and hence, martensite, at the grain boundaries of 430 stainless steel welds was eliminated completely when quenched with liquid tin. The weld solidification structure of 310 stainless steel, on the other hand, was essentially unaffected by quenching. Based upon the observations made, the weld microstructure of these stainless steels was summarized. The effect of cooling rate on the formation of primary austenite in 309 stainless steel welds was discussed. Finally, a simple method for determining the relationship between the secondary dendrite arm spacing and the solidification time, based on welding speeds and weld pool configurations, was suggested.     

An Investigation of the Massive Transformation from Ferrite to Austenite in Laser-Welded Mo-Bearing Stainless Steels

A series of 31 Mo-bearing stainless steel compositions with Mo contents ranging from 0 to 10 wt pct and exhibiting primary δ-ferrite solidification were analyzed over a range of laser welding conditions to evaluate the effect of composition and cooling rate on the solid-state transformation to γ-austenite. Alloys exhibiting this microstructural development sequence are of particular interest to the welding community because of their reduced susceptibility to solidification cracking and the potential reduction of microsegregation (which can affect corrosion resistance), all while harnessing the high toughness of γ-austenite. Alloys were created using the arc button melting process, and laser welds were prepared on each alloy at constant power and travel speeds ranging from 4.2 to 42 mm/s. The cooling rates of these processes were estimated to range from 10 K (°C)/s for arc buttons to 10⁵ K (°C)/s for the fastest laser welds. No shift in solidification mode from primary δ-ferrite to primary γ-austenite was observed in the range of compositions or welding conditions studied. Metastable microstructural features were observed in many laser weld fusion zones, as well as a massive transformation from δ-ferrite to γ-austenite. Evidence of epitaxial massive growth without nucleation was also found when intercellular γ-austenite was already present from a solidification reaction. The resulting single-phase γ-austenite in both cases exhibited a homogenous distribution of Mo, Cr, Ni, and Fe at nominal levels.     

Dynamic phase transformation during superplastic deformation of Nb/Nb<Subscript>3</Subscript>Al <Emphasis Type="Italic">in-situ</Emphasis> composite

Nb_ss/Nb₃Al in-situ composite with the nominal composition of Nb-16 mol pct Al-1 mol pct B, consisting of bcc niobium solid solution (Nb_ss) and A15 ordered Nb₃Al, was synthesized by arc melting, homogenization annealing, and isothermal forging, and their superplastic deformation behavior was investigated by tensile tests and microstructure observations. Maximum superplastic elongation over 750 pct was obtained at 1573 K and at a strain rate of 1.6 × 10⁻⁴ s⁻¹ for as-forged specimens. Phase transformation from Nb_ss to Nb₃Al was observed to occur during superplastic deformation. Dynamic phase transformation during superplastic deformation progresses more quickly than static phase transformation during annealing without applied stress. Dynamic phase transformation is accompanied by phase-boundary migration, which operates as an accommodation process of grain-boundary sliding. Dislocation creep dominates deformation and grain-boundary sliding is inhibited at a high strain rate, while grain-boundary sliding and cavity formation are promoted at a low strain rate because of insufficient accommodation of grain-boundary sliding arising from sluggish dynamic phase transformation. It is concluded that there exists an optimum strain rate that guarantees the grain-boundary sliding and the rapid dynamic phase transformation to achieve maximum superplastic elongation.     

Autogenous electrospark deposition of NiCoCrAlY

Abstract

Electrospark deposition is a microarc welding process that has been successfully used to deposit an autogenous ultrafine NiCoCrAlY coating. Primary solidification of the β phase shows epitaxial growth where orientation and heat flux coincide. Electron channelling contrast imaging of the deposits shows a homogeneous distribution of non-stoichiometric phases from the rapid solidification of the electrospark deposition process. Electron backscatter diffraction mapping of the deposit indicates that an ultrafine β structure can grow epitaxially where favourable conditions prevail. The equiaxed structure associated with the γ to β solidification shift was not observed and is believed to result from the homogeneous electrode and substrate compositions. Oxidation of the deposit shows advantageous alumina growth stemming from the ultrafine aluminium rich β phase. The transition from θ- to α-alumina was nearly complete in only 6 h at 1000°C in air.     

Time-resolved x-ray imaging of aluminum alloy solidification processes

Time-resolved direct-beam X-ray imaging, with intense, coherent, and monochromatic third-generation synchrotron radiation, and a high-resolution fast-readout detector system have been used for in-situ studies of dendritic and eutectic growth processes in Al-Cu alloys. Temporal and spatial resolutions down to 0.25 seconds and 2.5 μm, respectively, were obtained with a field of view up to 1.4×1.4 mm². Solid-liquid interfaces and various phase-specific segregates could be observed, and their dynamics could be traced in a sequence of temporally resolved images formed by phase and amplitude contrast from the sample. This article does not present any detailed analysis of a specific solidification phenomenon; instead, it presents to the scientific community an innovative technique for in-situ monitoring of such a phenomenon in real metallic systems.     

Solidification of an alloy 625 weld overlay

The solidification behavior (microsegregation, secondary phase formation, and solidification temperature range) of an Alloy 625 weld overlay deposited on 2.25Cr - 1Mo steel by gas metal arc welding was investigated by light and electron optical microscopy, electron microprobe, and differential thermal analysis techniques. The overlay deposit was found to terminate solidification at ≈ 1216 °C by aγ/Laves eutectic-type reaction. The Laves phase was highly enriched in Nb, Mo, and Si. The solidification reaction and microsegregation potential of major alloying elements in the overlay deposit are compared to other Nb-bearing Ni base alloys and found to be very similar to those for Alloy 718. Solidification cracks observed in the overlay were attributed to the wide solidification temperature range (≈170 °C) and formation of interdendritic (γ+Laves) constituent. Reasonable agreement is obtained between the calculated and measured volume percent (γ+Laves) constituent with the Scheil equation by treating the overlay system as a simpleγ-Nb “binary” and using an experimentally determinedk ^Nb value from electron microprobe data.     

Use of electrical resistance testing to redefine the transformation kinetics and phase diagram for shape-memory alloys

The phase-transformation temperatures of a nickel-titanium-based shape-memory alloy (SMA) were initially evaluated under stress-free conditions by the differential scanning calorimetric (DSC) technique. Results show that the phase-transformation temperature is significantly higher for the transition from detwinned martensite to austenite than for that from twinned martensite (or R phase) to austenite. To further examine transformation temperatures as a function of initial state, a tensile-test apparatus with in-situ electrical resistance (ER) measurements was used to evaluate the transformation properties of SMAs at a variety of stress levels and initial compositions. The results show that stress has a significant influence on the transformation of detwinned martensite, but a small influence on the R-phase and twinned martensite transformations. The ER changes linearly with strain during the transformations from both kinds of martensite to austenite. The linearity between the ER and strain during the transformation from detwinned martensite to austenite is not affected by the stress, facilitating application to control algorithms. A revised phase diagram is drawn to express these results.     

<Emphasis Type="Italic">In-situ</Emphasis> three-dimensional microstructural investigation of solidification of an Al-Cu alloy by ultrafast x-ray microtomography

This article presents the first results of a new experimental technique developed to investigate the evolution of the morphology of the solid and liquid phases during the solidification of a metallic alloy. It consists of ultrafast X-ray microtomography observations of a solidifying aluminum-copper alloy carried out at ESRF. These experiments allow investigating in-situ the formation of the casting microstructure and of the evolution of the morphology of the solid and the liquid phases. It allows also the in-situ determination of the solidification path, of the variation of the copper content in both the liquid and solid phases, and of some other characteristic parameters of the microstructure. Provided that some forthcoming technical improvements on the experimental setup are performed, more quantitative results can be obtained as well as better image quality and resolution.     

Kinetics of peritectic reaction and transformation in Fe-C alloys

In situ dynamic observation of the progress of a peritectic reaction and transformation of Fe-(0.14 pct C)- and Fe-(0.42 pct C)-peritectic Fe-C alloys has been successfully made with a combination of a confocal scanning laser microscope and an infrared image furnace. The peritectic reaction is characterized by the formation of the γ-austenite phase at the junction of the liquid and the grain boundary of δ-ferrite crystals and subsequent propagation of the three-phase point, liquid/γ/δ, along the liquid/δ boundary, whereas the peritectic transformation occurs by the thickening of the intervening γ toward both the liquid side and the δ side. The rates of the peritectic reaction for the two peritectic alloys are found to be much faster than the rate that would be controlled by carbon diffusion, suggesting that either massive transformation to γ or solidification as γ controls the rate. This is also the case for the Fe-0.14%C transformation in the hypoperitectic alloy. However, the rate of the peritectic transformation in the Fe-0.42%C alloy is determined by carbon diffusion. This article is based on a presentation made in the “Geoffrey Belton Memorial Symposium,” held in January 2000, in Sydney, Australia, under the joint sponsorship of ISS and TMS.     

Austenite Nucleation and Growth Observed on the Level of Individual Grains by Three-Dimensional X-Ray Diffraction Microscopy

Austenite nucleation and growth is studied during continuous heating using three-dimensional X-ray diffraction (3-D XRD) microscopy at the European Synchrotron Radiation Facility (ESRF) (Grenoble, France). Unique in-situ observations of austenite nucleation and growth kinetics were made for two commercial medium-carbon low-alloy steels (0.21 and 0.35 wt pct carbon with an initial microstructure of ferrite and pearlite). The measured austenite volume fraction as a function of temperature shows a two-step behavior for both steel grades: it starts with a rather fast pearlite-to-austenite transformation, which is followed by a more gradual ferrite-to-austenite transformation. The austenite nucleus density exhibits similar behavior, with a sharp increase during the first stage of the transformation and a more gradual increase in the nucleus density in the second stage for the 0.21 wt pct carbon alloy. For the 0.35 wt pct carbon alloy, no new nuclei form during the second stage. Three different types of growth of austenite grains in the ferrite/pearlite matrix were observed. The combination of detailed separate observations of both nucleation and growth provides unique quantitative information on the phase transformation kinetics during heating, i.e., austenite formation from ferrite and pearlite.     

The nonunjform distribution of inclusions in low-alloy steel weld deposits

Nonmetallic inclusions in low-alloy steel welds have an important effect on the microstructure and properties of weld deposits. This work is an attempt at understanding the factors controlling the spatial distribution of such inclusions, with particular emphasis on the uniformity of the distribution and the effect of solidification mode during manual-metal-arc welding. The solidification mode has been controlled by using unusual combinations of base plates and experimental electrodes. It is found that the first phase to solidify (in the form of columnar grains) is delta-ferrite (δ) when a medium carbon electrode is deposited onto a low carbon substrate, but that it is austenite (γ) when a low carbon electrode is deposited onto a high carbon substrate. Relatively large inclusions have been found to position themselves preferentially, during solidification, to the columnar grain boundaries of the first phase to solidify, whether this is 8-ferrite or austenite. The results can be understood qualitatively in terms of a surface tension driven Marangoni effect, or in terms of the pushing of solid inclusions by the solidification front. Both mechanisms drive the larger inclusions into cusps in the interface while smaller ones are passively trapped. The implications of the observed nonuniform distribution of inclusions are more severe for solidification with austenite as the primary phase, since the larger inclusions are in that case located in the weakest region of the weld where they also do not contribute to the intragranular nucleation of acicular ferrite.