The lecturer summarizes several steps in the application of thermodynamics to problems of ferrous metallurgy. His first effort in this field was a study of deoxidation of liquid steel. Other problems of interest in open-hearth operation included the carbon-oxygen relation, the influence of slags, absorption of hydrogen by the bath and the removal of sulfur.
Studies aimed at blast furnace problems included evaluation of thermodynamic activities of lime, silica, magnesia, alumina and sulfur in the slag.
A brief account is given of the lecturer’s contribution to the litigation in Detroit concerning the Austrian patent on the L.D. or B.O.F. process.
More recent work has been directed toward a better understanding of austenite, the high temperature form of iron which is of paramount importance in solidification, heat treatment and hot-working of steel. The activity of carbon in this solid solution has been determined as a function of composition and temperature and the effects of the important alloying elements, manganese, silicon, nickel, chromium, molybdenum and vanadium have been established. Some results are given of an as yet unpublished study of the system Fe–Mn–C and of the manganese carbides in steels. The lecture concludes with suggestions for further work on solid steels.

View full abstract

Ring-shaped platelets prepared by vacuum melting Ni–Fe–Ta alloys of 65.00∼83.65% Ni, 8.00∼24.70% Fe and 3.20∼23.90% Ta were heated in pure hydrogen atmosphere at 1150°C for 3 hr and then cooled at various rates. Measurements of magnetic properties, specific electrical resistivity and hardness of the specimens have revealed that the addition of Ta to Ni–Fe alloys greatly improves these characteristics. The highest values of initial permeability and maximum permeability obtained are 57300 and 428000, respectively, for the composition of 73.00% Ni, 12.05% Fe and 14.95% Ta when cooled at a rate of 800°C/hr from a temperature above the order-disorder transformation point. The coercive force and the intrinsic magnetic induction of the alloy are 0.0052 Oe (B_m=5000 G) and 7420 G (H=1000 Oe), respectively; its electrical resistivity and Vickers hardness are 64.3 μΩ-cm and 197, respectively.

View full abstract

The Young’s modulus at −150∼400°C and the elastic modulus, thermal expansion and hardness at room temperature have been measured for Mn–Cu–W alloys subjected to cold working and various heat treatments. It is shown that an anomaly associated with the antiferromagnetic\ ightleftarrowsparamagnetic transformation appears on the Young’s modulus vs temperature curves of the ternary alloys slow-cooled after heating at 900°C for 1 hr. The Young’s modulus at room temperature is at a little variance between the states of annealing, cold working and water quenching, but its value has a tendency to increase with increase in Mn or W content. The temperature coefficient of Young’s modulus differs greatly by the processes of annealing, cold working, water quenching and reheating after cold working or water quenching; the value undergoes a remarkable variation with composition, also. The presence of a large positive maximum on each of the temperature coefficient of Young’s modulus vs composition curves indicates the Elinvar characteristics of the ternary alloys. The rigidity modulus and its temperature coefficient at room temperature show a very similar variation to that of Young’s modulus and its temperature coefficient for the composition, heat treatment and cold working. In general the hardness of the ternary alloys undergoes a complicated variation within the range of 100∼660 by cold working, water quenching and heat treatment after cold working or water quenching as well as with composition. The corrosion resistivity is fairly good.

View full abstract

Measurements of Young’s modulus in the temperature range −150∼400°C and of rigidity modulus and hardness at room temperature have been made for Mn–Ni–Cr alloys subjected to various heat treatment and cold working conditions. The results indicate that the variation of Young’s modulus with temperature for alloys which were slowly cooled after heating at 950°C for 1 hr shows an anomaly associated with the antiferromagnetic\ ightleftarrowsparamagnetic transformation. The values of Young’s modulus at room temperature for the alloys subjected to annealing are higher than those for the alloys subjected to cold working, water quenching or reheating after cold working or water quenching. Moreover, these values have a tendency to increase with increasing nickel or chromium content. As does with alloy composition, the temperature coefficient of Young’s modulus differs greatly by the difference in treatment, viz. annealing, cold working, water quenching, and reheating after cold working or water quenching. A large maximum of positive sign is shown on the curves for the composition dependence of the temperature coefficient of Young’s modulus, thus indicating the Elinvar property. The room-temperature rigidity modulus and its temperature coefficient show changes with composition, cold working and heat treatment. These changes are analogous to those for Young’s modulus and its temperature coefficient. In general the hardness undergoes a complicated variation by alloy composition, cold working and heat treatment, the values being within the range of 100∼550 in Vickers hardness. Some of the ternary alloys have fairly high corrosion resistivity.

View full abstract

Dislocation motion in copper was studied by dislocation-etching before and under loading, and also after unloading.
Parallelepiped copper single crystals had a dislocation density of approximately 1×10³/cm² after thermal cyclic annealing. The dislocation etching under load was accomplished by hanging a small weight on the specimen in the etching solution.
Principal results are as follows:
(1) Dislocation locking stress due to impurity atoms was estimated to be less than 0.46 g/mm², assuming that impurity atoms act only as pinning points.
(2) Lattice frictional stress for dislocation motion was evaluated to be, at most, 0.20 g/mm², from the consideration of force balance on a bowed-out dislocation near the crystal surface.
(3) Both grown-in and multiplied dislocations were observed to move backward by unloading. Fresh dislocations made a back motion more easily.
(4) From an examination of the distance of dislocation motion and the average distance between grown-in forest dislocations, it is suggested that the main obstacle to moving dislocations is forest dislocations.

View full abstract

The EMF of an oxygen concentration cell in sulfide melts (matte and white metal) has been measured under a reduced pressure of around 0.5 mmHg at 1200°C by means of a solid-electrolyte cell and the results are presented as a function of the calculated partial pressure of oxygen, together with the EMF of molten slag, blister copper, and the Cu–O system. Zinc in matte affects the EMF to a great extent.
The magnetite contents in matte, white metal, and slag are semi-quantitatively determined by a magnetic method. The amount of SO₂ gas liberated during the vacuum treatment is discussed in relation to the difference of magnetite contents before and after the vacuum treatment.

View full abstract

Bilayered scale of NiO is formed on not only impure but high-purity nickel by heating in air at temperatures of 900 to 1200°C for times up to 72 hr. It is found that scales consist of two layers which are metallographically and/or crystallographically different from each other. They can be classified into several types according to their metallographic and crystallographic structures.

View full abstract

A survey is given on long period superstructures of close packed metallic phases comprehending shift variants of the Cu₃Au type, stacking variants of the Cu₃Au type and structures which display both variations. Shift variants and stacking variants obey valence electron rules for the shift density and h-density respectively. For the shift density a cut off interval exists in the neighbourhood of the valence electron concentration 1. The atom positions in the neighbourhood of the shift plane are somewhat changed and a possible rule for these changes is given. The shift normal is a direction of minimum elastic modulus following the presently known structures. Whole number half periods are preferred and give therefore a plateau of the shift density in dependence of the valence electron concentration. Nonintegral half periods are firmly established. There is a connection between the change of the axial ratio of the substructure and the change of the valence electron concentration.This axial ratio change is reduced by a second shift system. Shift variants occur with approximate valence electron concentrations 1 and 3. The rule of minimum iteration is seldom violated in shift variants and in stacking variants. For all these facts interpretations considering the spatial correlation of the electrons are given. To confirm these interpretations several open problems which are enumerated must be solved.

View full abstract

A survey is given on the structural family of beta brass related structural types. The members of this family are given by a vacancy forming process, by different possibilities of strains and by shifts. The multitude of structural types may be rationalized by a model which considers two spatial correlations of electrons, one for the valence electrons and the other for the outer core electrons.

View full abstract

The striation formation in copper crystals grown at the rate of 1 mm/min by the horizontal zone melting method was studied by a dislocation etch pit technique. It was found that the formation of striations depended upon the crystallographic orientation, the density and distribution of dislocations in a seed crystal and the axial temperature gradient. The [110] crystals showed a marked tendency to develop the striations, while the [111] and [100] crystals exhibited ordinarily a weak tendency of the striation formation. In the latter two crystals, the inoculation with seed crystals which were slightly damaged could result in growing the striated crystals, while no striations were formed when the seed crystals were not damaged. With a low dislocation density seed, the striations were not formed in the [111] crystal grown under a low axial temperature gradient, though a fine network of subboundaries appeared. On the other hand, the striations were formed after an incubation distance of 1 cm when the [111] crystal was grown under the same thermal conditions but with a high dislocation density seed crystal containing a fine network of subboundaries. However, under a high axial temperature gradient, no striations were built up, even if the crystal was grown with a high dislocation density seed crystal. Observations of the dislocation etch pits revealed that the striations wereformed when the coalescence of small angle tilt boundaries, which arranged closely in parallel, occurred.

View full abstract

Martensitic transfromation in TiNi alloys of the near-equiatomic composition has been studied by utilizing electron microscopy and diffraction under various thermal treatments. It has been observed that the martensitic transformation is associated with the variants of the product. Two orientations of the martensitic phase were observed as adjacent regions in quenched alloys. These orientations were forming in such a manner that they bore a common (1\bar1\bar1) plane and their b and c axes were interchanged. The slow cooled samples and those annealed at intermediate temperatures (650°C) showed (1\bar1\bar1) twinning in addition to the variants of the same transformation. The morphology of the (1\bar1\bar1) twinned product appeared as alternating platelets forming as adjacent regions. The samples that were aged at low temperatures (200°C) appeared to show internal twinning on (010) plane. The formation of these variants is considered in relation to the existing theories of the martensitic transformation. A possible mechanism of the shape memory phenomenon is also discussed.

View full abstract

In order to measure the effective diffusivity of gas in a porous solid, the permeability was measured by using the pressure gradient technique in the pressure range between 10⁻³ mmHg and atmospheric pressure. The specimen employed was a disc-shaped porous metallic molybdenum pellet prepared by the thermal decomposition of molybdenite pellet at 1290°C in vacuum.
Below the pressure of 1 mmHg, the measured permeability converges into constant values which are 1.98, 2.40 and 6.51 cm²/sec at 15°C for Ar, N₂ and He gases, respectively. These values correspond to the effective Knudsen diffusivities of each gas and they satisfy the relationship that the effective Knudsen diffusivity is inversely proportional to the square root of molecular weight of diffusing gas. From these results, the mean pore radius is calculated at \bara=1.11 μ according to the random pore model of monodisperse pore distribution.
The effective Knudsen diffusivity of diatomic sulfur gas at 1290°C in the same pellet was estimated at 3.60 cm²/sec. This value is in excellent coincidence with the estimated diffusivity of 3.52 cm²/sec which was calculated from the thermogravimetric data on the assumption that the diffusion of diatomic sulfur gas through the porous metallic molybdenum layer determines the overall rate of thermal decomposition of MoS₂ pellet.

View full abstract

Atomic ordering in the copper-platinum alloy system has been studied by X-ray diffraction using single crystals with compositions of Cu-48, 54, 61, 66 and 70 at%Pt. At the composition near the equiatomic one, an ordered structure CuPt, first found by Johansson and Linde, has been confirmed. A negative result has been obtained for the existence of an ordered structure Cu₃Pt₅, which was proposed by Linde. An ordered structure formed at the composition near CuPt₃ has been shown to be consistent with the structure model of Tang rather than that of Schneider and Esch.

View full abstract

An austenitic 18–8 stainless steel was tested in tension under hydrostatic pressure and the influence of hydrostatic pressure on its stress-strain behaviour was discussed. Tensile tests under constant pressures of 12000 kg/cm² and atmospheric pressure and tests during which the ambient pressure was changed from 12000 kg/cm² to atmospheric pressure and vice versa were carried out at room temperature. Also, the change in incuctance of a coil wound around the specimen was measured to evaluate the amount of α-martensite induced by plastic deformation. The results obtained are as follows: (1) Pressurizing at 12000 kg/cm² gives no effect on the stress-strain relation of the annealed specimen at atmospheric pressure. (2) In the magnetic measurement, α-martensite cannot be observed over the range of less than about 10% strain, whereas at a larger strain it begins to be induced and its amount increases similarly with increasing strain under both 12000 kg/cm² and atmospheric pressure. But ε-martensite is observed by X-ray diffraction when the specimen is deformed under 12000 kg/cm². (3) When the specimens are deformed under a constant pressure of 12000 kg/cm², the increase in flow stress against that at atmospheric pressure is very large at the beginning of deformation, decreases with increasing strain, and then becomes almost constant (about 12%) above about 15% strain or more. (4) When the ambient pressure is changed between atmospheric pressure and 12000 kg/cm² on the way of deformation, the flow stress changes 0 to 4% at small strains and, about 12% (pressure raising) and 19% (pressure releasing) at 30 to 40% strain. (5) The phenomena mentioned above in (3) and (4) can be explained by the fact that ε-martensite is induced rapidly at the beginning of deformation resulting in work-hardening under high hydrostatic pressure, while it easily transforms to α-martensite under a low tensile stress at atmospheric pressure at large strains.

View full abstract

The particle coarsening in Fe-20 wt% Cu alloy, in which the solid particles are bonded to each other by a number of interparticle welds, has been studied in the temperature range between 1100 and 1250°C. The other structural properties that change with particle coarsening, such as the size and the mean free path of solid particles, the average size and the number of interparticle welds were determined by quantitative microscopic analyses for the specimens sintered at 1150°C. The mechanism of particle coarsening is analyzed on the basis of the experimental results obtained. The particle coarsening can not be explained by the Lifshitz-Slyozov-Wagner theory alone. The increase in the size of the interparticle welds and the decrease in the number of them suggest that solid particles coalesce as a consequence of the weld growth. The kinetic equation of the growth of the interparticle welds during particle coarsening is derived, and it is proposed that the particle coarsening occurs substantially due to the coalescence of solid particles. The discontinuity in size distributions of solid particles is also discussed in terms of the coalescence hypothesis.

View full abstract