A coupled electrical field-circuit method was introduced to simulate the control process of electrical resistance welding. The circuit method was used to analyze the response of the control system and the field method was adopted to describe the behavior of the welding joint. By using a coupled electrical-thermal-mechanical FEM technology, the relationship of electrical, thermal and mechanical field was established. The coupled FEM analysis is adopted to analyze the behaviors of the welding joint. An equivalent circuit was introduced to combine the dynamic electrical resistance of workpieces, the welding power source and the control characteristics of the system. An updated method was selected to describe the dynamic response of the control system. In every increment, the variables were updated as initial conditions and boundary conditions in the field analysis, or resistance, voltage, current in the circuit analysis. An example was selected as projection welding process.

View full abstract

A program of research was carried out to investigate the dependence of laser welds in aluminium alloys upon proces parameters. Four different laser systems were used: carbon dioxide (CO₂), carbon monoxide (CO), pulsed neodymium yttrium aluminium garnet (Nd:YAG) and high powered continuous Nd:YAG lasers.
This paper describes phenomenon that occur at the surface of the weld pool and keyhole.
The energy transfer from the laser beam to the workpiece, termed the coupling is a crucial factor in determining the heat input to the weld which is one of the major process variables. The physical mechanisms involved in this multi-part transfer are complex, hence an experimental route was taken to determine the coupling and its dependence upon laser and material properties. Coupling was measured for the combinations of five aluminium alloys and four laser systems.
The high surface temperatures found in laser welding result in preferential evaporation of certain alloying elements. By studying welds made with a range of laser systems, the dependence of this effect upon process parameters could be observed. The high evaporation rates were also found to be linked to another problem that arises in laser welding-that of undercutting in autogenous welds.
Porosity was observed and is discussed here in terms of its mechanism of formation.

View full abstract

The purposes of this study are to perform the welding in a microgravity environment and to investigate the differences in welding phenomena such as the bead shape, the grain structure, the Vickers hardness and the microstructure between in a terrestrial environment and in a microgravity environment. A small sized electron beam welding apparatus was developed and aluminum-copper alloys, A2219, were welded using the developed apparatus under microgravity of approximately 10^-5G. As a result, it was found that a stable electron beam was obtained in both terrestrial and microgravity environments. A much flatter weld bead was formed in the microgravity environment than in the terrestrial environment. The grain size in the middle part of the weld bead was smaller in the microgravity environment than in the terrestrial environment. However, the Vickers hardness were similar in both terrestrial and microgravity environments.

View full abstract

The effects of welding conditions, tool rotation speed, Rt and plate traveling speed, V on weld joints formation and mechanical properties have been investigated on Friction Stir Welding joints of high strength aluminum alloys of 2024-T6 and 7075-T6 in comparison with 5083-O of 4 mm thick plates. Rt/V, Welding parameter, which is closely related to welding heat input per unit length of the welded joint affected joint qualities. Peak temperature of FSW joints near stir zone during welding increased with increasing Rt/V, but lower than solidus temperature for each alloy. As weld defects, lack of bonding and inner defect occurred at low Rt/V for each alloy, and excess high Rt/V caused surface tearing as defect both for 2024-T6 and 7075-T6. As to a range of optimum welding condition, 5083-O was wider than those of 2024-T6 and 7075-T6, and optimum Rt/V range common to three alloys was 3.3 to 5.0. In case of any aluminum alloy, tensile strength of the welded joints increased with increasing Rt/V, and weld defects caused low tensile strength. The maximum tensile strengths for 2024-T6, 7075-T6 and 5083-O were 373, 444 and 296 MPa in as-welded condition, which were almost 76, 83 and 100% to base metal strength. Post-weld artificial aging increased the tensile strength of 7075-T6 joint to 512 MPa, 90% to base metal strength, but in 2024-T6 decreased it slightly.

View full abstract

The nitrogen absorption by iron, Fe-20Cr-10Ni and SUS329J1 stainless steel YAG laser welding in the atmosphere of Ar-N₂ mixture gas was investigated in comparison with those during arc welding using the same materials as in this experiment and equilibrium data. Although the nitrogen contents of YAG laser weld metal increase with the nitrogen partial pressure were as well as those of arc weld metal of arc welding, the nitrogen content during YAG laser welding were quite less than those during arc welding. Blowholes can not be observed in Fe-20Cr-10Ni and SUS329J1 stainless steel and can only be found in iron at lower nitrogen partial pressure during YAG laser welding. A discussion on the difference in nitrogen absorption between YAG Laser and arc welding has suggested that small amount of nitrogen absorption results from less opportunity of nitrogen to touch the surface of molten metal due to the active evaporation of metal which covers the major surface of molten metal during laser welding metal. Additionally, the short-time thermal cycle compared with arc welding may bring insufficient nitrogen absorption in the weld metal during cooling. It can be considered that the nitrogen absorption during YAG laser welding is basically different from that during arc welding.

View full abstract

Joing of Si₃N₄/Si₃N₄ was carried out with a newly developed Cu₆₈Ni₁₀Ti₂₁B₁ brazing alloy. With the powdered new brazing alloy, the shear strength of the joints reaches 233 MPa. With the rapidly-solidified foils of the brazing alloy, the three-point bend strength of the joints is raised to 406 MPa at room temperature and this value is kept until 723K. The interfacial reactions of the joint and its effects on the joint strength are discussed. The lattice matching between Si₃N₄ and Ti₅Si₃ is better than that between Si₃N₄ and TiN. So the formation of Ti₅Si₃ rather than TiN at the interfaces is favorable to the joint strength. Adding element Ni to Cu-Ti alloy can weaken the activity of Ti in the alloy, which is favorable to decrease the thickness of interfacial reaction layer and in the result the joint strength is increased. An adding a small amount of B in the new alloy can decrease its liquidus temperature.

View full abstract

When many additional alloying elements are used for high strength of steels, the weldability is usually poor. The development of high-strength steels of 800 MPa class with good weldability is preferred to proceed on the basis of a new concept for innovative materials design which may remarkably reduce the carbon equivalent of steels. The higher-strength steels of 700 to 800 MPa class with the same carbon equivalent as mild steels could be achieved by creating ultra fine ferrite grains of less than 1μm.
Efficient and high speed joining processes with low heat input and narrow width of heat affected zone (HAZ) were developed to obtain high-performance welded structures while preserving ultra fine microstructure. The softened region in the HAZ of ultra fine-grained steels can be controlled with the newly developed welding process.

View full abstract

Due to the influence of welding thermal cycle toughness of structural steel degenerates. Recently, the intercritically reheated coarse grained zone (IC CG HAZ) was found display the worst toughness in welded joint, which was associated with its fracture mechanism. In this work, two IC CG HAZs of a structural steel were prepared by welding thermal-cycle simulation techniques. For the two IC CG HAZs the dynamic fracture toughness and the fracture mechanism were studied, and the correlation between fracture toughness and fracture mechanism was also discussed. Under both static and rapidly loading, cracks in the IC CG HAZ were found to initiate at the intersection of α_B⁰ packets with different orientations, followed by propagating in cleavage. In some places of crack propagation, adjacent cleavage facets are connected by shear, producing dimple zones. Though the brittle fracture initiation mechanism remains unchanged, the cleavage facet size and the proportion of the dimple zones between facets vary with loading speed. These changes are shown being associated with the effect of strain rate on fracture toughness. In this work, the fracture behavior of IC CG HAZ was also compared with that of base metal and prestrained steel.

View full abstract

The materials used were cast alloy 718 plates with three different grain size levels. The plates were heat-treated at 1368K×5.4ks or 1378K×10.8ks before welding. The results of the longitudinal Varestraint test indicated that the fine-grained and heat-treated (1378K×10.8ks) specimens were considerably less sensitive to HAZ cracking. Based on the microscopic observations of specimens heated at various temperatures, the grain boundary liquation mainly occurred in the laves cluster which had a low melting point. In addition, the degree of grain boundary liquation in the fine-grained and heat-treated (1378K×10.8ks) specimens were less than that of the coarse-grained and non heat-treated specimens. The EPMA analysis revealed that S was severely segregated at the laves cluster in the interdendritic region. The amount of S segregation decreased with decrease in grain size or homogenization heat treatment. A theoretical calculation suggested that the elevated grain boundary liquation temperature was caused by decreasing the S segregation in the laves cluster. On the basis of these results, the HAZ cracking susceptibility of cast alloy 718 could be reduced by decreasing grain size or homogenization heat treatment before welding. The decreased HAZ cracking susceptibility was considered to be attributed to the reduction in grain boundary liquation due to decrease in the amount of laves cluster and S segregation which resulted in an increased grain boundary liquation temperature.

View full abstract

Effect of simulated welding cycles consisting of austenitizing at 1623K/6s and cooling for Δt_8/5=6s (the 1^st cycle), reheating at 873-1123K/6s and cooling for Δt_8/5eff=6 or 40s (the 2^nd one), and annealing at 473-893K/6s and cooling for Δt_8/5eff=6s (the 3^rd one) on the evolution of microstructure and changes in absorbed energy (Charpy test at 293K) of SQV-2A low alloy steel was investigated. Light microscopy, TEM, electron diffraction, SEM, EDX, and measurement of hardness HV10 were used to characterize the microstructure of the steel including type of observed phases and morphology of fracture surfaces. The formation of M-A constituent and its internal evolution during 2^nd and 3^rd welding cycles (i.e. precipitation and growth of carbide particles) as well as the morphology of intragranular interlath M-A islands were found to be responsible for changes in absorbed energy. The role of M₃C and M₂C carbides in influencing the absorbed energy seems to be more significant than suggested earlier.

View full abstract

A study was conducted to examine the effects of postweld heat treatment (PWHT) on the impact toughness and microstructures in the weld heat affected zone (HAZ) of Cu-containing HSLA-100 steel. The Gleeble, thermal/mechanical simulator, was used to simulate the weld HAZ. The details between toughness and PWHT of HAZ were studied by impact test, optical microscopy (O.M.), scanning electron microscopy (SEM), transmission electron microscopy (TEM) and differential scanning calorimetry (DSC). The decrease of HAZ toughness in the single thermal cycle simulation relative to the base metal's toughness is ascribed to the coarsened-grain formed by heating to 1350°C. While the increase of HAZ toughness in the double thermal cycle simulation relative to that in the single thermal cycle is due to the fine ferrite (α) grain transformed from austenite (γ) by heating to α/γ two phase region. Cu precipitated during aging, which increases the strength of the base metal, is dissolved during the single thermal cycle to 1350°C and remains in the matrix during subsequent thermal cycle simulations. It precipitates by introducing PWHT suggesting that the decrease of toughness in the triple thermal cycle simulation (T_P1=1350°C, T_P2=800°C and T_P3=500°C) does not occur from the Cu precipitation. The behaviors of Cu-precipitates in the weld HAZ are similar to those in the base metal. The PWHT at 550°C shows the highest hardness and the lowest toughness in the HAZ, whereas PWHT at 650°C shows a reasonable toughness and strength combination suggesting the possibility of improving the toughness of HAZ.

View full abstract

A majority of cooling system failures in many corrosion resistant alloys is around weldments. Weld regions are particularly attractive to microbes as welding alters the material surface characteristics. Hence, it is presumed that there is a correlation between the susceptibility to this type of corrosion and the microstructure. Existing literature hardly mentions anything about the preferential adhesion of bacteria on areas of varying microstructures. One of the important characteristics of weld is its microstructure. A study involving both microstructure and bioadhesion would reveal the reason why welds suffer preferential MIC attack. Experiments were carried out to study the effect of microstructure on adhesion of a corrosive bacterial strain Bacillus sp. isolated from the residual water of an MIC affected effluent treatment plant on weld samples (weld metal, HAZ and base metal separately) of two different materials viz. 316L and 304L stainless steel.
Base metal, HAZ and weld metal of both the material tested, showed difference in area of adhesion. Weld metal or HAZ harboured more bacteria in both the materials tested, with base metal showing the lowest. A difference in percentage area of adhesion was observed between as welded and polished coupons of the same material. Since base metal, HAZ and weld metal of both the materials showed difference in area of adhesion in spite of the uniform surface condition, the influence of microstructure gathers significance. This preferential adhesion contributes very much to corrosion and can be considered as one of the factors causing MIC attack on welds.

View full abstract

The effect of the substrate properties, namely temperature and position, on the deposits formed by the new process of plasma spraying of submicron ceramic suspensions using a continuous ink jet printer was investigated. The substrate temperature was shown to have a considerable effect on the structure of deposits formed. At low substrate temperatures the splats produced were irregular and poorly shaped. Preheating the substrate to a temperature above 400°C produced circular splats. Such dependence of splat morphology on the substrate temperature was compared with that observed for conventional plasma spraying. Splats formed by the new method were an order of magnitude smaller than conventionally sprayed splats and splats smaller than several microns were found to be free from microcracking. Overspraying to form coatings showed that when the substrate was far from the plasma (low substrate temperature) a poorly densified coating was produced. Placing the substrate closer to the plasma (hotter substrate) resulted in a more coherent and denser coating being produced. These effects were explained in terms of the spreading and cooling rate experienced by the molten drops as they impinge on the substrate.

View full abstract

In this study, the influence of substrate effect on the evaluation of mechanical properties of thin films was investigated by making nanoindentation tests on the CrB thin films on the different substrates. These results show that the critical contact depth/thickness ratio (D/t ratio) for determine the true hardness of the thin films is 0.06, which is smaller than the suggested value of 0.1 by ASTM and UK National Physical Laboratory. In the case hard film on the hard substrate, the tested hardness value decreases with increasing D/t ratio above 0.06 due to the pile-up of film materials next to the indenter. In the case of hard film on the soft substrate, the tested hardness value decreases with increasing D/t ratio above 0.06 because deformation zero extends to the interface and the substrate due to the low hardness value of the substrate. Although the Young's modulus values of the films keep constant below the critical contact depth/thickness ratio (D/t ratio) of 0.06 in this study, they show a great difference with different film substrate systems. These results show the rule that CrB film on the substrate, which has a high Young's modulus, shows high Young's modulus and no clear trend with hardness of substrate is apparent.

View full abstract

The paper presents a study on the obtaining of NiAl and Ni₃Al refractory intermetallics in their stoichiometric and homogeneity range, obtained by self-propagation high temperature synthesis, the thermo-explosion variant. The processing variables were the compacting pressure and the material chemical composition. The influence of unreacted materials green porosity on the sintered materials contraction/expansion was determined. The interdependence of the liquid amount appeared during synthesis of the reaction products and the green as well as final material porosity were established.

View full abstract

The paper presents a simple method to recycle the bearing steel scrap in view to recover the solid part and to use it as additive in sintered friction materials. The method consists of treating and removing of oils, water and surface oxide film. The recovered powders consist of a mixture of metallic and nonmetallic particles. The amount and the chemical composition of the non-metallic part vary with the type and the quality of the grinding wheel. The recovered powder mix was used as such or in the composition of some friction materials, in a proportion of 80 wt%. The obtained results suggest the recovered powders can be used in the following fields: 1) metallic parts as limit-switcher or spacers; 2) sintered friction materials; 3) friction additive for the organic matrix friction materials.

View full abstract

Dense TiAl/Ti₂AlC composites were produced from mixed powders of Ti, Al, and TiC by pulsed-electric current sintering. The sintered products mainly consisted of TiAl and Ti₂AlC phases and showed different microstructures depending on the composition. When 7vol%TiC was mixed in the starting powders, the Ti₂AlC particles were uniformly dispersed in the Ti-Al matrix with the duplex structure containing γ and lamellar phases. When 15vol%TiC was used, the product was composed of an interpenetrating network of the Ti₂AlC and TiAl phases. After the heat treatment, however, both microstructures changed to that of Ti₂AlC particles dispersed uniformly in the matrix of a γ single phase. The fracture toughness, flexural strength, and hardness of these composites as well as the wear resistance and friction coefficient were evaluated.

View full abstract

A technique for optimization of steel chemistry for MnS precipitation on oxide inclusions has been developed through employing the computational thermodynamic method. The optimum composition of Si/Mn deoxidized steel was estimated to be Fe-0.1mass%C-0.1mass%Si-1.5mass%Mn-0.01mass%O-0.007mass%S. The prediction was validated by experiments using a semi-levitation melting technique. The inclusions mainly consisted of MnO-SiO₂ oxides with MnS precipitates. The observed morphology revealed a possibility of both oxide and sulfide having been precipitated as liquid phase, although the existing thermodynamic data do not support this to occur.

View full abstract

Thermodynamics of the MnO-SiO₂-TiO₂ and the MnO-Al₂O₃-TiO₂ systems have been studied for optimizing the property of oxides in steel. Iso-activity curves of each component, which are necessary for the determination of the composition of steel equilibrated with such oxides, have been estimated in the wide range of liquid phases. Although in the study of the MnO-SiO₂-TiO₂ system, iso-activity curves of SiO₂ and TiO₂ were partially obtained due to the lack of integration limit on the saturated line of SiO₂ and TiO₂, integration limits of each component have been derived by ternary Gibbs-Duhem equation. This enabled the estimation of steel composition in equilibrium with a wide range of the oxide system. In addition, the behavior of sulfur in the MnO-SiO₂-TiO₂ and the MnO-Al₂O₃-TiO₂ oxides in steels were simulated using the sulfide capacities of these systems in order to optimize processing conditions.

View full abstract

For the continuous casting of aluminum containing stainless steel, there should be a minimal amount of silica in the mold powder. It oxidizes aluminum in molten steel, allowing chemical composition to vary, and undesired inclusions to form. In mold powder, silica is important component to lower the melting point and maintain adequate viscosity. A substitute for SiO₂ might be TiO₂, because of its nature as an acidic compound. Although the liquidus and solidus of the system is minimum knowledge for the melting point, they have not been well established.
The phase relations of the Al₂O₃-CaO-TiO₂ system were investigated by the hot filament technique under various oxygen partial pressures. The effect of B₂O₃ addition to the Al₂O₃-CaO-TiO₂ system on the liquidus and solidus was also investigated.

View full abstract

The solubility of carbon in the MO-B₂O₃ (MO=CaO, BaO, and Na₂O) and CaO-RO (RO=SiO₂ and Al₂O₃) slags at high temperatures was measured to understand the thermodynamic behavior of carbon in molten slags. The solubility of carbon as a function of the composition of slags shows a minimum value, indicating that carbon dissolves by different mechanisms in the acidic and basic slags, respectively. The infrared spectra measurements indicate that the B-C bond is about 1150 and 1140 cm^-1 in the acidic region of the CaO-B₂O₃ and Na₂O-B₂O₃ slags; hence, the incorporation of carbon into the borate network is confirmed qualitatively. The dissolution of carbon and nitrogen into the slags shows a similar behavior in the B₂O₃-bearing slags, while different behavior is shown in the CaO-SiO₂ and CaO-Al₂O₃ slags, based on the relationship between carbide capacity and nitride capacity.

View full abstract

The behavior of MgO-C refractory-slag-metal system, which is caused by the reactions such as the dissolution of MgO and graphite in the refractory into slag and metal respectively and the generation of gas bubbles, was observed directly by using a high temperature X-ray radiographic apparatus and analysed theoretically. The local corrosion is regarded as due to the cyclic dissolution of MgO and graphite in the refractory into slag and metal phase respectively. The local corrosion is greatly influenced by the generation position of gas bubbles and the number of gas bubbles generated. Bubbles generated in the refractory-slag-metal three-phase boundary restrain the local corrosion, while bubbles generated at the refractory-metal interface enhance the local corrosion. Gas bubbles form mainly according to the reaction between (FeO) in the slag film and C_(s) in the refractory contacting with the slag film: (FeO)+C_(s)=Fe_(l)+CO_(g). Some fundamental principles to improve MgO-C refractory were put foward.

View full abstract

There is a limit for lowering nitrogen in steel because of pickup from atmosphere during and after degassing. Therefore, nitride capacity of various fluxes has been measured in order to examine the possibility of nitrogen removal by using slag. It is necessary to take into account the nitrogen reaction rate between gas and slag phases and know the nitrogen behavior through successive phases of metal, slag, and gas for the efficient removal of nitrogen in practical steelmaking process. In this study, nitrogen removed rate from molten iron-aluminum alloy to gas phase through CaO-Al₂O₃ melts has been examined and a mixed rate determining model has been applied through successive phases of metal, slag, and gas taking gas/slag reaction rate into consideration. For the effective removal of nitrogen from metal to gas phase through CaO-Al₂O₃ melts, it is important to keep oxygen pressure in the gas phase higher and oxygen potential at slag/metal interface lower.

View full abstract

An expression for the accurate prediction of the viscosities of various industrial slags has been derived using a model based on the concept of network structure. The equation is expressed in terms of slag components and some basic physical quantities. The viscosity predictions closely fit with the experimental data for a large number of blast furnace type slags.

View full abstract

As a part of investigation on the scrap-base steel making, a computer model, which could simulate the transport phenomena in heating of a steel cylinder in a graphite vessel, was developed. The model describes the numerical modeling of two-dimensional coupled turbulent flow and heat transfer in a graphite vessel. The model uses generalized transport equations, which are applicable to the gas and solid region. The governing transport equations are solved simultaneously by the ANSYS package. This model allows calculation of the temperature distribution inside the gas and solid phase, as well as convective and radiative heat transfer coefficients. Launder and Spalding's default values have been used with satisfactory results. The effect of various factors, such as steel cylinder surface conditions, gas flow rate, on the heating rate has been discussed. The values of emissivity for different surface condition have been determined. Best fitting was obtained with an emissivity's value of 0.9 for black surface and 0.2 for the polished one. The radiative heat transfer coefficients obtained were 560 (for polished surface) and 640 W/m²K (for black surface), respectively. The convective heat transfer coefficients were in the range of 13 to 33.64 W/m²K, depending on the gas flow rate. The convective Nusselt number in dependence on the Reynolds and Prandtl numbers yielded to the following relation: Nu=0.75Re^0.446Pr^0.33.

View full abstract

The methods available for the measurements of the following thermophysical properties of high temperature melts are reviewed viz. density, viscosity, surface tension, heat capacity, enthalpy, thermal diffusivity and conductivity, electrical conductivity and emissivity. The methods are illustrated with reference to work carried out on molten silicon. In order to combat contamination of the melt from container/sample reactions there has been a gradual switch to the use of (i) containerless methods employing various forms of levitation or microgravity (drop towers, parabolic flights and space experiments) and (ii) sub-second rapid heating and other non-intrusive techniques. However, there are other equally-important conditions which must be addressed to obtain accurate property values, e.g. the control of oxygen partial pressure in surface tension measurements and the selection of the most reliable method to analyse data. It is our contention that equal attention should be given to these problems.

View full abstract

The determinations of the various physical properties of silicon involved in fluid flow and heat transfer (during single crystal growth) have been collated and critically evaluated. Recommended values for silicon are given for the following physical properties and their temperature dependencies: density, viscosity and surface tension, heat capacity, enthalpy, thermal and electrical conductivities, thermal diffusivity and emmisivity.

View full abstract

Theoretical models describing the transport of oxygen in metal/atmosphere systems under different fluid-dynamic conditions have been developed by different authors. In the present study, as in previous ones, the molecular diffusion is the process mainly controlling the exchange of matter between the liquid metal and the atmosphere. So, in this paper a diffusional model is proposed accounting for volatile oxides and for gas phase homogeneous reactions by means of two limiting conditions: instantaneous reactions and null reactions. For the boundary conditions, the model assumes a bulk flow composition of the gas layer surrounding the liquid on the upper side and a local equilibrium constraint at the liquid interface. The asymptotic behavior of the system is described, enabling the prediction of the direction of the net oxygen flux. It has been demonstrated that the results obtained are valid for any type of homogeneous gas phase reactivity, provided that no oxide fog is formed in the gas. The model is useful to correctly guide technological processes such as single crystal growth: in the paper the application to the melt silicon/oxygen system is discussed. Finally, the present model can synthesize apparently contradictory experimental measurements of surface tension available in literature into a unique portrait.

View full abstract

The surface tension of molten Si was measured using the electromagnetic levitation method under a microgravity condition. The experiments were carried out over a wide temperature range between 1460K (below the melting point) and 1880K. The atmosphere was Ar+3%H₂ purified by Platinum asbestos and magnesium perchlorate. The oxygen partial pressure is estimated to be at least than 1.1×10^-14Pa. The shape of the droplet is controlled by changing the output ratio for two different coils. The equilibrium shape of the oscillating droplet was precisely determined by calibrating the apparent change in the droplet shape. The effect of the droplet distortion on the shift in the oscillation frequencies and the calculated surface tension value was also investigated. When the equilibrium shape of the levitating droplet is spherical, the surface tension of molten Si was obtained with very small scatter. The value is expressed as follows: γ=-0.074(T-1687)+735γ: Surface tension (mN/m), T: Temperature (K).

View full abstract

Dependence of the surface tension of molten silicon on temperature, the oxygen partial pressure in ambient argon atmosphere have been determined at temperatures ranging from 1693 to 1873K and in the range of P_O2 from 10^-25 to 10^-14 MPa using a high purity BN substrate with the sessile drop method. Change of the surface tension with the oxygen partial pressure is very small in the range of P_O2≤10^-22 MPa. The surface tension decreases remarkably with increasing the oxygen partial pressure in the range of P_O2 from 10^-22 to 10^-20 MPa, and increases slightly with increasing the oxygen partial pressure in the range of P_O2>P_O2, sat. The temperature coefficient of surface tension, δσ/δT, is minus and increases with increasing P_O2, δσ/δT at P_O2=10^-25 MPa is equal to about -0.74 mN·m^-1·K^-1 and then δσ/δT gradually increase with increasing the oxygen partial pressure up to 10^-22 MPa. δσ/δT steeply increases with increasing the oxygen partial pressure in the range of P_O2 from 10^-22 to 10^-20 MPa. And again gradually increases with further increases of P_O2 up to 10^-15 MPa, where δσ/δT is equal to -0.15 mN·m^-1·K^-1. It is found that the surface tension of molten silicon and its temperature coefficient are sensitive to the oxygen partial pressure of argon atmosphere in the measuring system.

View full abstract

oxygen partial pressure in ambient atmosphere has been considered to affect the surface tension of molten silicon and its temperature coefficient. However, the effect of the oxygen partial pressure on the Marangoni convection, which is driven by the surface tension gradient on a silicon melt, has not yet been made clear through experimentation. The temperature oscillation due to Marangoni convection in the half-zone bridge molten silicon (5 mm in height; 10 mm in diameter) was measured by fine thermocouples in 1-G experiments. We conducted X-ray flow visualization with tracer particles in ambient atmosphere at a variety of oxygen partial pressures in argon gas under microgravity using the parabolic flight of an aircraft. The mode transition from an oscillatory flow with multiple frequencies to that with a single frequency took place when the oxygen partial pressure was increased to P_O2=1.8×10^-5 MPa. Also, the tracer velocity of the Marangoni convection clearly decreased with increasing oxygen partial pressure under microgravity. These experimental results can be explained by the dependence of temperature coefficient of surface tension of molten silicon on oxygen partial pressure of ambient atmosphere.

View full abstract

Surface tension measurements have been carried out on binary alloys of iron and silicon, and on pure iron using the levitated drop technique. The alloy compositions (atomic percent) chosen were Fe+25% Si, Fe+37.5% Si, Fe+50% Si, Fe+62.5% Si, Fe+75% Si. Attempts were also made to measure the surface tension of pure silicon by this method, which unfortunately resulted in only one successful measurement. Surface tension is derived from measurements of the oscillation frequencies of a liquid droplet levitated in the magnetic field of a radio frequency coil, which also inductively heats the sample. It is therefore necessary for the material under study to be electrically conducting, and whilst the alloy samples levitated readily, it was necessary to pre-heat the silicon to about 800°C before it would levitate. The difficulties the electrical conduction of silicon imposes on the measurement are discussed. In order to produce high purity samples, with low oxygen contents, samples were prepared in a cold crucible levitator under an argon +10% hydrogen atmosphere.
The surface tension values obtained in this study were:

Fe  1.89-0.44×10^-3 (T-1537 [°C]) Nm   O content <5 ppmFe+25% Si  1.69-0.19×10^-3 (T-1263 [°C]) Nm  O content 10 ppmFe+37.5% Si  1.51-0.24×10^-3 (T-1247 [°C]) Nm  O content 5 ppmFe+50% Si  1.17+0.19×10^-3 (T-1407 [°C]) Nm  O content <5 ppmFe+62.5% Si  1.02-0.14×10^-3 (T-1298 [°C]) Nm  O content 10 ppmFe+75% Si  0.87+0.24×10^-3 (T-1242 [°C]) Nm  O content 31 ppm

A single value for the surface tension of silicon of 0.78 Nm at 1900°C was obtained.

View full abstract

The capillary flow rate through the porous glass media under microgravity was measured by using the drop-shaft type microgravity facility of Japan Microgravity Center. Based on a simple model, the rate of the capillary rise rate of liquids was developed and expressed by
t=(8μ/a²ρg)(h₀ln(h₀-h))-h)
h₀=2σ cosθ/aρg
where t is time, μ is a viscosity coefficient, a is a capillary radius, ρ is a density, g is the gravitational constant, h is a capillary rise distance and h₀ is the equilibrium rise distance.
The behavior of the capillary rise in the porous media under microgravity is reasonably explained by the developed equation. But, the capillary rise along the vertical direction under normal gravity did not agree with the equations, and this disagreement was discussed based on the dynamic contact angle. It is estimated that the dynamic contact angle under μG will be different from that under 1G.

View full abstract

Structure evolution and deformation behavior of a 304 type austenitic stainless steel were studied in multiple compression at a temperature of 873 K (0.5 T_m) under a strain rate of about 10^-3 s^-1. The integrating flow curve shows a maximum at moderate strains around 1.5 followed by a minor strain softening at high cumulative strains above 3. The structural changes taking place during deformation can be characterized by the evolution of elongated subgrains with their low-to-middle angle boundaries as dense dislocation walls at low to moderate strains. These subgrains become more equiaxed and the subboundary misorientations gradually increase with increase in strain. The volume fraction of the highly misoriented substructure substantially increases upon multiple deformation to high cumulative strains above 3, finally leading to the development of a fine grained microstructure with an average grain size of about 300 nm. The mechanisms of such structural development as well as the relationship between microstructures and deformation behavior are discussed in details.

View full abstract

Grain refinement of 304L stainless steel was examined by mechanical milling. Fine austenitic microstructure materials with grain sizes smaller than 1μm were obtained by consolidating the mechanically milled powder with nano-meter-sized martensitic microstructure and reversing to austenite during the hot isostatic pressing (HIP). It is important for grain refinement to form finer martensitic microstructure with random crystallographic orientations before the reversion treatment. The strongest material showed the microstructure with the average grain size of 0.3μm and 3 times of 0.2% proof stress and 1.5 times of tensile strength compared with conventional steel. Hot forging after HIP treatment was effective to improve the ductility because the interfaces between powders were strengthened. The forged materials had elongation over 30% and reduction of area over 50%. The strengthening of as-HIP materials is mainly effected by grain refinement, because the dependence of proof stress on grain size agreed with the Hall-Petch relationship. M₂₃C₆ on grain boundaries and nano-meter-sized precipitates in grains, identified as Cr-carbite or Si-oxide, were observed in the microstructure. These precipitates might be related not only to suppressing the coarsening but also to strengthening. The dislocation structure was observed in the microstructure of the forged materials. The strengthening of the forged materials is probably controlled by work-hardening and recovery during forging process.

View full abstract

Microstructures and mechanical properties were investigated in ultra fine-grained oxide-dispersion steels with 24 mass% Cr ferrite matrix fabricated from mechanical milled powders. Powder mixtures of Fe, Cr and Y₂O₃ were mechanically milled in an argon gas atmosphere and then consolidated at around 1100K. After 720 ks mechanical milling (MM), nanocrystalline ferrite grain structures were formed within powder particles. As for Y₂O₃ particles. It was observed that they were not only crushed to several nanometers but also decomposed during MM, and then oxides like Y₂O₃ and YCrO₃ finely precipitated in ultra fine grained structure on annealing at around 1200K. Such ultra fine oxide particles effectively retarded the growth of matrix grains. The sample with 1 vol.% Y₂O₃ addition maintained ultra fine ferrite grains of 0.4 μm even after the annealing of 1573 K-3.6 ks by the Zener pinning effect of the oxide particles of about 10 nm in diameter. Tensile strength of the bulk samples markedly rose as the ferrite grain size became smaller in the grain size range below 1 μm and reached 1.93 GPa for the bulk sample with 1 vol.% Y₂O₃ addition and with the smallest mean grain size of 0.16 μm.

View full abstract

Ultra rapid heat treatment is a means of obtaining annealed products with improved productivity, whilesaving space cost and energy. It may also produce new time-temperature-microstructure relationships and still unknown effects on surface, microstructure and textures. Therefore some properties are expected to be modified when ultra rapid heat treatments are applied to standard stainless steels grades like AISI 304 and AISI 430.
Before considering such a process suitable to be applied industrially, it is necessary to establish the effect of rapid heat treatment on the metallurgy of the particular stainless strip products and their relevant formability performance which is of paramount importance for their acceptability on the market.

View full abstract

A High carbon martensitic stainless steel (Fe-12%Cr-0.7%C) was subjected to isothermal aging after full solution treatment, and then solution-treated again in the two phase region of austenite and M₂₃C₆ carbide; partial solution (PS) treatment. Microstructural development during these heat treatment was investigated, and mechanical properties for the PS treated steel were evaluated by comparison with a referencial steel without isothermal aging. During the isothermal aging at 923K, the supercooled austenite undergoes eutectoid transformation and decomposes to ferrite and M₂₃C₆ carbide, and the carbide particles are finely dispersed within the matrix. Insoluble carbide particles retained during the following PS treatment contribute to suppressing austenite grain growth effectively through the grain boundary pinning effect. Ductile-brittle transition in the PS treated steel is characterized by a lower transition temperature and a higher upper-shelf energy compared with the referencial steel. This is due to the grain refining of prior austenite as well as the homogeneous dispersion of carbide particles through the PS treatment with isothermal aging.

View full abstract

Direct quenching (DQ) has become widespread instead of conventional reheat quenching in order to obtain good mechanical balance of high strength steels in addition to productivity, energy savings, alloy savings, and weldability. It is well known that mechanical properties of DQ steel can be changed by varying slab reheating temperature, finish rolling temperature, and cooling rate in addition to alloy composition. The effect of pancaking by controlled rolling at low temperatures on mechanical properties was investigated. The effect of pancaking depends on hardenability determined by alloy contents. Pancaking increased strength or toughness when alloy-determined hardenability is high. But pancaking decreased strength when alloy-determined hardenability is low.

View full abstract

The dislocation substructure in compressively deformed fcc 70Ni-30Fe and 67Ni-30Fe-3Ti alloys was investigated to understand or model microstructural evolution in austenite of low carbon steels during hot deformation. These Ni-based alloys are expected to show similar deformation characteristics to austenitic steels because of similar stacking fault energy. The deformation substructure depended in particular on deformation temperature. Below 700°C a heavy deformation (70% reduction) introduced microbands, while above 800°C it produced equiaxed dislocation cell structures with some partially recrystallized austenite grains. Crystallographic misorientation in microband structures tended to be larger than that in cell structures at the fixed strain and temperature. Preexisting precipitates effectively suppressed the growth of austenite grains recrystallized after deformation, which resulted in macroscopically homogeneous dislocation substructures consisting of microbands. The temperature dependence of dislocation substructures may closely relate to the mechanism of diffusional transformation to ferrite at very low temperature 500°C through strain assisted transformation in low carbon steels.

View full abstract

In order to study the interface boundary structure, a Near-Coincidence Site Lattice (NCS) theory based on Co-incidence Site Lattice (CSL) has been proposed. In this method, coincidence atoms are calculated by a computer based, the atomic matching procedure. The NCS analysis has applied to BCC/HCP systems in three-dimensions to predict results of TEM observation in a Ti-22V-4Al alloy. As the results of HRTEM observation of an α precipitate in β grain, an α precipitate has a lath shape: a broad face boundary consist of misfit dislocations whose spacing is irregularly, while, a side facet boundary are made of regular structure ledges. The NCS analysis is able to predict the α/β interface boundary structure using lattice parameter and orientation relationship between α precipitate and β matrix, and shows good correspondence with the experimental results of side facet obtained by HRTEM observations. Therefore, it is suggested that atomic matching affects strongly the low energy interface boundary structures and the NCS analysis is a useful method to predict the structure of interface boundary.

View full abstract