Hafnium carbide (HfC) whiskers were grown on a Ni/graphite substrate from a gas mixture of HfCl₄, CH₄ and H₂ at temperatures between 1273 and 1673 K. The maximum growth rate was 0.278 μm/s in length and 2.78 nm/s in diameter. The preferential growth direction was ⟨110⟩. The lattice constant of the HfC whiskers was (0.4641±0.0001) nm, which corresponds to the carbon to hafnium atom ratio, C/Hf, 0.94±0.02. The HfC whisker was found to grow by the VLS mechanism at temperatures higher than 1573 K, while at temperatures lower than 1573 K by the VS mechanism.

View full abstract

The microstructure and the crystalographical coherency of plated chromium films on copper substrate have been investigated by transmission electron diffraction and microscopy. The orientation relationships between a deposited chromium film and copper substrate are
(This article is not displayable. Please see full text pdf.)
\ oindentFor instance, when the crystal is in the relationship of [112]Cr\varparallel[110]Cu, the difference between the lattice spacing of (112)Cr plane and the lattice spacing of (1\bar10)Cu plane which are along the direction of [1\bar10]Cu is −8.54% and it is not so large. But atom arrangement of the chromium atoms and of the copper atoms along this direction is not in good condition to fit each other, consequently misfit strain exists in the interface between the deposited chromium film and the substrate. And in the transmission electron micrographs of chromium film with the relationship shown above, many micro twin crystals are observed along the [110]Cu and [1\bar10]Cu direction on the relationships of a vertical direction each other. On this investigation it has been found that these many micro twin crystals contribute to reliefe this large misfit strain. So these micro twin crystals in the chromium film would be misfit twin crystals.

View full abstract

When an n⁺ type surface of a diode is bonded to an electrode using Al solder, a thin p-type Si layer, i.e. a regrowth layer is formed on the n⁺ type Si after bonding. Because of the layer formation, a considerably high forward voltage drop (FVD) is observed in the regrowth layer. In order to reduce the FVD, a brazing mechanism between n⁺ type Si evaporated with Al and W electrode evaporated with Cu has been studied in the temperature range from 800 to 873 K. The Cu content was varied from 12.4 to 62.3 wt% by varying the thickness of Cu film. The FVD and the regrowth layer are compared with the previous case of Al solder without Cu film.
The brazing temperature between n⁺ type Si and W electrode was reduced to about 800 K. The temperature reduction was caused by the formation of ternary eutectic melt of the Al-Cu-Si system.
The FVD was substantially lowered in comparison with the case of Al solder. In order to prevent the crack formation in the brazed layer, it is necessary to keep the Cu content up to 29 wt%, corresponding to the eutectic composition in the Al-Cu-Si system.

View full abstract

Significant contribution of crystal lattice dislocations to the grain boundary sliding has been suggested by various experiments and theories. However, the difference between the behaviors of grain boundary sliding with and without crystal slip has been rarely reported.
In the present paper, the behaviors of grain boundary sliding are investigated by the use of orientation controlled cadmium bicrystals for the two cases mentioned above.
The rate of grain boundary sliding without crystal slip varies in proportion to the applied stress and is controlled by the grain boundary diffusion. When crystal slip is activated, the grain boundary sliding rate is increased. In such cases, the rate of grain boundary sliding is found to vary in proportion to the square of applied stress, and the activation energy for the grain boundary sliding is almost equal to that of volume diffusion. These results indicate that the mechanism of grain boundary sliding with crystal slip is different from that without crystal slip.
McLeans’ model for the grain boundary sliding with crystal slip is developed by the consideration that the number of dislocations in the grain boundary is proportional to the shear stress on the slip system. It is possible to explain variation of the grain boundary sliding rate accompanying the change in the shear stress on slip systems by this model.

View full abstract

Thermodynamic quantities of W₂B and WB were determined by the emf measurements of the cells with yttria-stabilized zirconia as the electrolyte: ⟨W⟩W, W₂B, B₂O₃|YSZ| air ⟨Pt⟩ and ⟨W⟩W₂B, WB, B₂O₃|YSZ| air ⟨Pt⟩. The standard free energies of formation of W₂B and WB are expressed by the following equations: ΔG_W2B^°=−97250+1.787T±900 J/mol (1100-1400 K) and ΔG_WB^°=−87840+1.652T±700 J/mol (1100-1400 K). A comparison of these results with the literature values has also been discussed.

View full abstract

In order to investigate the shape effect of the particles on the high-temperature strength of dispersion-hardened alloys, two kinds of nickel alloys with SiO₂ particles of complex and spherical shapes were made by internal oxidation and a heat-treatment after the internal oxidation, respectively. Their strengths were compared by means of hardness test at high temperatures.
The micro-Vickers hardness of the alloy with particles of complex shape was found to be always higher than that of the alloy with particles of spherical shape at all test temperatures from room temperature to 1273 K, but the difference in hardness depended on test conditions: while for a short loading time, only a slight difference was observed, the difference at high temperatures increased gradually with the increase in loading time and approached to a final constant value.
The dispersion hardening estimated from the final value was in good agreement with the Orowan stress calculated from the dispersion, which means that the final difference between the two alloys can be well explained by the difference in the particle density on a slip plane, and any shape effect on the decrease in high-temperature strength due to the local climb of dislocations around particles was not detected.
Further, the creep behaviour of the alloys was estimated from the loading time dependence of the hardness, and it was found that both of the stress exponent and the apparent activation energy for the creep were larger in the alloy with complex particles than in the alloy with spherical particles.

View full abstract

Static flow stress decreases abruptly when ultrasonic oscillatory stress (17.8 kHz) is intermittently superimposed. The decrement in static flow stress, Δσ was measured in an Al-5 at%Mg alloy. The ranges of strain rate and testing temperature were from 1×10⁻⁴ to 1×10⁻³ s⁻¹ and from 473 to 673 K, respectively. Δσ was zero at 673 K. This implies that dislocation velocity at 673 K is proportional to effective stress. Δσ at 473 and 573 K was dependent on oscillatory stress amplitudes, strain rates and testing temperatures. These experimental results indicate that the dislocation velocity-effective stress exponent, m^* is greater than unity at lower temperatures. The change in m^* with temperature and strain rate is satisfactorily explained in terms of a solute-dislocation interaction.

View full abstract

In order to make clear the mechanism by which CaO brings about the acceleration of the reduction of wustite, the dependences of the reduction rate of wustite containing CaO on the temperature and α-γ transformation of metallic iron were investigated. For the purpose, pellets of pure wustite and wustite containing CaO in several quantity levels were reduced by hydrogen over the temperature range from 873 to 1273 K. The following results were obtained.
(1) Wustite grains are covered by dense layer of metallic iron and intercepted from the reducing gas, when pure wustite is reduced at a higher temperature than 1073 K. The addition of CaO to wustite, however, can yield the metallic iron of such the property that should involve the fine pores open to reducing gas and wustite at any temperature. The fine pores in metallic iron, nevertheless, diminished gradually with increasing the temperature in the range of α-iron. However, when temperature rises from α-iron to γ-iron, the reduction rate increases discontinuously, probably because the sintering of fcc-iron is slower than that of bcc-iron. The above mentioned phenomena can be shown also by the change in the interfacial chemical reaction rate constant k_r which is estimated by analysing the reduction rate on the basis of the mixed control equation.
(2) The activation energy of the interfacial chemical reaction hardly change by the addition of CaO to wustite. This means that the reduction accelerating effect of CaO may not be catalytic, but it will be caused by the formation of iron of such a property that the direct contact between wustite and gas should be always maintained even under the coexistence of metallic iron.

View full abstract

The effects of non-stoichiometries on the oxidative leaching reactions of heazlewoodite samples in nitric acid solutions were examined from the kinetic point of view, carrying out the determination of leaching rates, the measurement of electric resistivity and so on.
Main results obtained are as follows:
(1) The oxidative leaching rates of heazlewoodite samples in nitric acid solutions were influenced greatly by the stoichiometries of the ores. Reaction with nickel-rich heazlewoodite proceeds faster than that with nickel-deficient ones.
(2) The electric properties of heazlewoodite samples were not influenced by the change in the stoichiometries of the ores. Therefore, the electric properties of heazlewoodite do not normally play a significant part in oxidative leaching reaction.
(3) The oxidative leaching reaction of heazlewoodite is a simultaneous reaction of sulfate ion formation and elemental sulfur formation.
(4) The overall reaction of oxidative leaching of the ores in nitric acid solutions is controled by a chemical reaction step, and the rate equation is given by the following equation.
(This article is not displayable. Please see full text pdf.)

View full abstract

A Co-0.45 mass%Ni alloy oxidizes parabolically by formation of a (Co, Ni)O scale when exposed in the oxygen pressure range from 2.6×10² to 10⁵ Pa at temperatures of 1373, 1473 and 1573 K. The nickel gradient within the scale decreased with increasing outward distance due to nickel migrating slowly than cobalt, which profiles tended to be flattened with decreasing oxygen pressures and elevating temperatures.
A computer simulation is made for the scale growth on the alloy and a cobalt metal. Computation demonstrates that agreement between the calculated and experimental determinations of the parabolic rate constants k_p was good, but not very good for dependences of values for k_p and concentration profiles upon the ambient oxygen pressures.
It was found at temperature of 1473 K that the oxygen activities (1.5∼4.1×10⁻³) shifted toward higher values than the equilibrium value of 2.9×10⁻⁵ at the oxide/alloy interface enable the calculated values for k_p and concentration profiles to coincide with the observed ones. These results could be ascribed to the formation of fine pores and micro-cracks in the vicinity of the scale/alloy interface. The analogous oxidation behavior might be expected to see for the scale growth on a pure cobalt metal.

View full abstract

A Cu-13.1 at%Be-0.2 at%Co crystalline alloy was directly produced by rapid solidification and then aged. Rapidly solidified foils about 250±50 μm in thickness were obtained by an piston-anvil type apparatus. The structure changes are investigated by means of micro-Vicker’s hardness (Lood, 2.94N), X-ray diffraction and electron microscopy. The rapid-solidification (R.S.) induces a low maximum hardness and prolongs the time necessary to attain the maximum hardness below 723 K, contrary to our experimental results for Cu-Ti⁽¹⁾ and Ni-Ti⁽²⁾ alloys. The Johnson-Mehl kinetic modes (n)⁽³⁾ of aging are about one for the R.S. alloy and about two for a commercially produced (C.P.) alloy below 723 K. The precipitation mechanism of the R.S. alloy is constructed by the preferencial fine precipitation of γ phase instead of γ′ phase. The dominant factor of age-hardening for the R.S. alloy is mainly the dispersion hardening mechanism of fine γ phase particles instead of the precipitation hardening of γ′ phase for the C.P. alloy.

View full abstract

An investigation of diffusivity and solubility of hydrogen in both annealed and as-cold-rolled specimens of an austenitic Type 304 steel has been carried out at room temperature by means of an electrochemical permeation method. Especially, the effect of formation of the “foreign phases”, such as hydride phases (γ_H, ε_H) and hydrogen-induced martensitic transformation ones (ε+α′) in the cathodic surface on the hydrogen permeation behavior is discussed. Results obtained were as follows:
(1) The linear relation of log(t^1⁄2·J_t) vs 1⁄t in the hydrogen permeation transients was valid at the beginning of the permeation process. The results were explained by a model in which the rate-determining step of the permeation is a hydrogen-diffusion process in the remaining γ-phase matrix of annealed specimen or in the remaining (γ+α′) phase matrix in the case of as-cold-rolled specimen, accompanying the phase transition of the γ or (γ+α′) phase matrices to the γ_H, ε_H hydride phases and/or hydrogen-induced transformation (ε+α′) martensite phases during permeation.
(2) The apparent diffusion coefficient of hydrogen determined from the theoretical permeation transient derived under the condition of the homogeneous medium throughout the specimen is equivalent to the true diffusion coefficient in the remaining matrix phase. However, the apparent hydrogen concentration directly beneath the cathodic surface determined by the theoretical solution of a homogeneous medium corresponds to the phase equilibrium hydrogen concentration of the matrix phase in the phase boundary related to the moving rate parameter of the phase boundary. Therefore, in order to determine strictly the concentration of hydrogen in the matrix phase at the interface of the foreign/matrix phase, especially in the case of higher cathodic current density for hydrogen introduction, it should be desired to use the heterogeneous medium solution related to the moving rate parameter of the interface.
(3) The diffusivity (D) and apparent solubility (C) of hydrogen in the annealed specimen obtained in the temperature range from 311 to 332 K, at i_c=50 A·m⁻² can be described as follows:
D(m²·s⁻¹)=(4.41±0.92)×10⁻⁷exp[−53510±750(J·mol⁻¹)⁄RT] and C(mol·m⁻³)=(6.11±2.17)×10⁵exp[−10170±1050(J·mol⁻¹)⁄RT].
These values were in agreement with those previously obtained from the gas [phase method at high temperature. While, compared with the annealed specimen, the as-cold-rolled one had the higher diffusivity and the low activation energy for hydrogen diffusion, due to the pre-existence of cold worked α′-martensite. The apparent solubility of hydrogen in the as-cold-rolled specimen was higher than that in the annealed one and the hydrogen absorption occurred by the exothermic reaction.

View full abstract

Forsterite ceramic-to-nickel and -to-Fe-Ni alloy seals were formed by a reactive alloying method utilizing titanium foil, and the seal strength was measured by tensile tests. Moreover, the interfaces between the materials were examined by optical microscopy, scanning electron microscopy, electron probe microanalysis and X-ray diffraction.
The best thermal treating condition is 1273 K as the heating temperature and 300-600 seconds as the holding time for this method. The highest seal strength was 5 MPa for the forsterite ceramic-to-nickel seal and 7 MPa for the forsterite ceramic-to-Fe-Ni alloy seal. The major part of the fractured surface in the samples sealed with the best condition was the ceramic itself. The lower the heating temperature and/or the shorter the holding time than these of the best condition, the more the separation was apt to take place at the interfaces between the titanium foil and the ceramic, while the higher the heating temperature and/or the longer the holding time, the larger the fractured surface at the interfaces between the titanium foil and the metal became.
Titanium diffused into the ceramic and metal, and the Ti-Ni binary alloy or the Ti-Ni-Fe ternary alloy was formed in the interfaces between the titanium foil and the metal. Titanium is considered to react with SiO₂ to form intermediate layers at the interfaces between the titanium foil and the ceramic.

View full abstract

The dry wear characteristic, hardness and bending strength were examined for composites with randomly oriented carbon fiber-chops in aluminium produced by the powder metallurgical process. It was possilbe to fabricate composites with fiber contents up to 30 vol%. The large reduction in the bending strength with fiber amounts was considered to be due to the large porosity (∼15%) and lack of the strong bonding between fibers and aluminium matrix.
It appeared that composites containing fiber-chops more than 20 vol% showed excellent wear resistance compared with that of a sintered aluminium, especially in the initial stage of the wear process. By the optical and scanning electron microscopes, the smooth cross section of fibers and few scratches on the aluminium matrix were observed for the worn surface of the composite. On the other hand, the well developed scale-like structure and heavily marked striations were revealed for a sintered aluminium. The good wear characteristic of the composite may be caused by the higher loading capacity and wear resistance of carbon fibers.

View full abstract

A study was conducted to determine the effects of microstructure of the carburized layer on the fatigue strength of Cr-Mo steel (SCM 420).
The vacuum carburizing process was adopted to change the microstructure of specimens, because the process was capable to control exactly the surface carbon content and the quenching temperature.
The reason of difference in the fatigue strength of the specimens which have various microstructures was discussed in term of the retained austenite, carbide and residual stress in the carburized layer.

View full abstract

Displacement of the data-point on the thermoelectric power versus electrical conductivity plot gives information about the segregation of carbon atoms to dislocations in alpha iron. From such a plot for cold rolling of a low-carbon aluminium-killed steel, it is deduced that the amount of carbon segregated to dislocations is larger after rolling at 360 K than after rolling at 290 K or 196 K. The concentration of carbon atoms occupying the hypothetical dislocation sites in deformed alpha iron on annealing for long times at 523 K is argued to be higher than the value predicted from a simple model. The relief of lattice strains by the segregation of carbon atoms to dislocations is manifested by the displacement of the data-point with a high gradient on the thermoelectric power versus conductivity plot. The gradient characteristic of the segregation of carbon atoms to dislocations is 8.0±0.7 μV·μΩ·m·K⁻¹.

View full abstract