The structure and electronic, magnetic and thermal properties of Nd–Fe and Nd–Fe–B amorphous alloys are modified appreciably by hydrogen charging. The feature and extent of the modification differ significantly between the two systems. This difference is primarily attributed to the effect of interatomic bonding, specifically Fe–B and Nd–H chemical bonding, as well as short-range order structure in the amorphous alloys. Changes in structure and electronic and magnetic properties of hydrogenated amorphous alloys have been investigated by measuring X-ray diffractographs, X-ray photoemission spectra, magnetic hysteresis curves and thermal desorption spectra, and discussed in terms of the bonding characteristics and SRO structure of the amorphous alloys.

View full abstract

A two-dimensional idealized nucleation and growth (2D-ING) model is considered, in order to understand the experimental results relating to martensitic phase trasition. It is an extension of the Kakeshita model for nucleation process (Kakeshita et al.: Mater. Trans., JIM, 34 (1993), 423), based on the nucleation probability derived from a nucleation barrier. We extend the Kakeshita model to an ideal growth process by introducing the concept of a dynamic embryo and a frozen nucleus. A dynamic embryo is a “non-classical” nucleus in the non-equilibrium state. After the size of an embryo is over a critical size, the embryo is transformed into a frozen nucleus. Domains in the low-temperature phase are assumed to develop gathering frozen nuclei. The results of a computer simulation based on the above model show the presence of an incubation time, which is one of the essential properties of the first-order phase transition, and display the cooperative formation of domains and the fractal distribution of their size. The experimental results on the kinetics of the martensitic phase transition in In-Tl alloys have been interpreted in terms of the above simulation results.

View full abstract

Grain boundary segregation of P and B in extra-low carbon cold-rolled steel sheet containing from 0.01 and 0.15 mass% P and up to 38 mass ppm B after annealing at 1123 K was investigated by Auger electron spectroscopy (AES). AES measurements on the intergranular fracture surfaces of the specimen indicate that the Auger peak intensity ratio (PIR) of P increased and that of B decreased with increasing bulk concentration of P, and that PIR of P markedly decreased, while that of B increased with a B addition of up to 10 ppm. In the bulk concentration range of B of more than 10 ppm, however, these ratios hardly changed. The segragation behavior was nearly the same between specimens annealed for 20 s and 500 s, indicating that the segregation of P and B was in an equilibrium state. Calculations of the grain boundary coverages of P and B indicate that (1) the segregation free energy of B was about twice that of P, (2) the coverage of B was larger than that of P in the specimen with B concentration of more than 10 mass ppm, and (3) the total coverage of P and B was nearly constant in the specimen with B concentration of more than 10 mass ppm. It is concluded that the improvement in secondary workability of P-added steel sheet by also adding B is attributable to the segregation of less P and more B on the grain boundary, resulting in enhanced grain boundary cohesion.

View full abstract

In order to obtain much information about the microscopic mechanism of martensitic transformations, ⁶³Cu nuclear magnetic resonance (NMR) and relaxation measurements have been performed at temperatures above and below the transformations in Au–Cu–Zn alloys with different compositions and transformation temperatures. The ⁶³Cu NMR spectra taken from the parent phase with a L2₁ type structure of the alloys comprising less and more Au contents can be decomposed into three and two components, respectively. It implies that the majority of Cu atoms occupy the corner sites, referring to the fundamental bcc lattice, and the minority of Cu atoms the body-centered sites with different Knight shifts. The component spectra reveal different distribution of the Cu atom, changing with Cu content. Although the NMR spectra taken from the martensite phase are too broad to decompose, we can obtain the average Knight shift with respect to the Cu atom distribution. The temperature dependence of the ⁶³Cu spin-lattice relaxation time follows the Korringa-like behavior in different ways by the alloys. The data on the relaxation rate are discussed in terms of the Knight shift value and the enhancement of electron-electron correlation. The transformation temperature has no clear relation with the NMR parameters.

View full abstract

A new nonequilibrium structure consisting of nanoscale amorphous particles surrounded by fcc-Al phase was found to form in the Al₉₄V₄M₂ (M=Fe or Co) and Al₉₃V₅Ni₂ alloys rapidly solidified at the circumferential velocity of 40 m/s. The deviation of the alloy component and the decrease in the circumferential velocity of the melt-spinning roll results in the formation of nanoscale mixed structures of fcc-Al plus icosahedral (I) phases or the fcc-Al plus I plus amorphous phases. The sizes of the amorphous and fcc-Al phase regions are about 10 nm and 7 nm, respectively, for the Al–V–Fe alloy and about 25 nm and 20 nm respectively, for the Al–V–Co alloy. The formation of the nanoscale amorphous particles in coexistence with the Al phase is presumably due to the suppression of the transition of supercooled liquid to I-phase resulting from the retardation of the diffusivity of the solute elements. The highest tensile strength reaches as high as 1390 MPa for the mixed phase Al₉₄V₄Fe₂ alloy and decreases significantly by the transition to I plus fcc-Al phases. The high tensile strength is presumed to result from the formation of the nanoscale mixed structure consiting of nanoscale amorphous and Al phases in which the Al phase has enriched solute concentrations and contains a high density of interphase boundaries. The first success of fabricating the nanoamorphous structure is particularly important for the future development of nanophase materials.

View full abstract

Simultaneous evaporation of Al (99.999% in purity) from dual sources of W baskets was carried out onto a surface of NaCl crystal cleaved in an ordinary high vacuum (10⁻³ Pa), and the crystallinity of the deposited film was examined through a transmission electron microscope (TEM). The substrate temperature varied in the range from room temperature to 523 K during the preparation of Al films of thickness 100 nm. On the other hand, the deposition thickness was changed from 20 to 130 nm at a substrate temperature of 473 K. A continuous single-crystal film with the (111) plane of thickness more than 70 nm was obtained on a substrate heated to 473 K.

View full abstract

Secondary recrystallization of an Fe-3%Si alloy containing AlN and MnS as inhibitors with heavy cold rolling before primary recrystallization annealing was investigated with special reference to poor secondary recrystallization.
Excessive cold-rolling reduction produces smaller primary recrystallized grains with higher intensity of {100}⟨025⟩ oriented grains and higher possibility for {100}⟨025⟩ oriented grains associated with Σ5 coincidence boundaries. Specific grain growth of the {100}⟨025⟩ primary recrystallized grain before the onset of secondary recrystallization is considered to cause the poor secondary recrystallization by reducing the driving force of secondary recrystallization. It may be essential for this unstable secondary recrystallization that the two equivalent orientations of the {100}⟨025⟩ oriented grains which increase with increasing cold-rolling reduction have the Σ5 coincidence orientation relationship with each other.

View full abstract

Martensitic transformations which occur in Ti-45.5Ni-5.0Cu and Ti-44.0Ni-5.0Cu-1.0Al (at%) alloys were investigated by means of electrical resistivity measurements, X-ray diffraction analysis and transmission electron microscope observation. It was found that both the two alloys transform in two steps on cooling, i.e., from the high temperature cubic phase (B2-type structure) to the orthorhombic martensite (B19-type structure), and then to the monoclinic martensite (B19′-type structure). The two-step transformations in these alloys proceed as follows. On cooling, plates of orthorhombic martensite nucleate at certain sites in the cubic matrix. While the plates of orthorhombic martensite are in growth, plates of the monoclinic martensite nucleate in the orthorhombic martensite and grow consuming it. The transformation from the orthorhombic martensite to the monoclinic martensite occurs by shearing on (001)_o plane to [100]_o direction. A plate of the monoclinic martensite which have grown in this way consists of thin plates of (001)_m compound twin as a result of alternative shearing to [100]_o and [\bar100]_o direction. It is difficult to differentiate the successive two transformations of these alloys described above by electrical resistivity measurement or X-ray diffraction because the two transformations overlap and the amount of the orthorhombic martensite in the specimen through the two-step transformation is very small.

View full abstract

Hot extrusion has been carried out at 573, 673 and 753 K on machined chips of an AZ91 magnesium alloy and mechanical properties of the extrusions have been investigated. The extrusions processed from machined chips showed a good combination of high ultimate tensile strengths of 320∼410 MPa, high 0.2% proof stresses of 220∼300 MPa and elongations to failure of 5∼12% at room temperature. Grain refinement and dispersion of oxide layers by hot extrusion were responsible for the good mechanical properties at room temperature. Also the extrusions processed from machined chips at 573 and 673 K showed superplastic behavior at 573 K, though the elongations of the extrusions processed from machined chips were lower than those of the extrusion processed from the cast ingot.

View full abstract

Calcium fluoride solid electrolyte has been extensively used in galvanic cells for the measurement of fluorine potential to obtain the activity of alloys and compounds, because the ionic transference number of the electrolyte is about unity even at P_F2⁄P⁰=10⁻⁶² above 550°C. So far there were no literature data available for the measurement below 500°C by use of the calcium fluoride solid electrolyte.
The purpose of this work is to develop the sensors for the control of aluminum concentration in a hot dip galvanizing bath, which is of industrial importance for producing high quality galvannealed steel sheets.
In this report, a trial has been made to measure the fluorine potential in a molten zinc bath artificially equilibrated with ZnF₂ at the temperatures between 450∼500°C using CaF₂ cells with some kinds of reference electrode. By use of Bi, BiF₃ or Zn, ZnF₂ as the reference electrode, the cells respond quickly and stable emf values which agree fairly well with the literature data are obtained.
Subsequently, a trial has been made to measure the fluorine potential in molten Zn–Al baths artificially equilibrated with AlF₃ at the temperatures between 450∼500°C using CaF₂ cells with the above reference electrodes. The initial response of these cells after immersion of the cells into these baths is relatively slow. However, once emf of the cell becomes stable, each cell responds quickly to the aluminum concentration change. The response time is not longer than 5 min. From the analysis of such emf measurements, the emf values are found to increase linearly with the logarithm of the aluminum concentration. The possibility of the present method has been ascertained in Zn–Al baths.

View full abstract

This report describes the electrical conductivity measurements for metal sulfides containing Be, Mg, Ca, and Sr in group IIa and Sc and Y in group IIIa. To avoid contamination, preparations of sulfide disks were strictly controlled.
All the group IIa metal sulfides showed predominantly ionic conduction. Comparison of the measured activation energies for these metal sulfides with those for other alkaline-earth sulfides and oxides indicated that the predominant charge carriers were the metal cations in these ionically conductive sulfides. At higher pressures, the conductivities of MgS and BeS were linearly dependent on the sulfur partial pressure showing a slope of 1/6 on a logarithmicplots. This indicates that these sulfides are p-type electrical conductors in that pressure range. A theoretical analysis showed that fully ionized metal vacancies are the predominant defects. For Sc₂S₃ and Y₂S₃, thisslope was −3⁄16, indicating predominance of scandium and yttrium interstitial ions. At higher pressures, Y₂S₃ also showed a +3⁄16 dependence, indicating p-type conduction due to fully ionized yttrium vacancies.

View full abstract

Iron distribution in thin film oxide on silicon wafers is investigated in conjunction with the generation of a negative charge due to Fe in the oxide. The investigation uses the ac surface photovoltage (SPV) technique. Prior to thermal oxidation, Si wafers are treated with an Fe-contaminated alkaline solution composed of ammonium hydroxide, hydrogen peroxide, and water. During thermal oxidation, Fe atoms which are deliberately added to the wafer surface tend to segregate at the top of the oxide layer. This is confirmed by a sophisticated impurity-extraction technique combined with a high-resolution chemical analysis technique. For Fe-contaminated oxide films on n-type Si wafers, ac SPV appears, indicating that the negative charge due to Fe in the native oxide survives during thermal oxidation. The negative charge disappears after etching off the surface layer of the oxide to a depth of about 7 nm. In p-type Si, the negative charge compensates the positive fixed oxide charge localized at or near the SiO₂–Si interface, resulting in the reduction of the ac SPV. The creation of the negative oxide charge is attributed to the replacement of quadrivalent Si ions by trivalent Fe ions in the oxide.

View full abstract

The liquid-crystalline transformation behavior during continuous cooling and the transformation-induced structure were examined for a Zr₆₀Al₁₀Ni₁₀Cu₁₅Pd₅ molten alloy which was ejected into a wedge-shape cavity in a copper mold. The wedge-shape cavity has a constant depth of 50 mm and different vertical angles (θ) ranging from 5 to 15 degrees. The ejection temperature of the molten alloy was also changed in the range of 1273 to 1573 K. The cast structure consists only of a glassy phase in the θ range smaller than 10 degrees and changes to a mixed structure consisting of glassy and nonequilibrium crystalline Zr₂Ni and Zr₂Cu phases in the higher θ range. The glass transition temperature and crystallization temperature of the cast metal glass are 683 and 778 K, respectively, which agree with those for the melt-spun glassy ribbon. The start (Cs) and termination (Ct) points for the transformation from the supercooled liquid to crystalline phases during continuous cooling were determined from the thermal analytical data obtained at different sites in the wedge-shape cavity and the continuous-cooling-transformation (C.C.T.) curves were constructed. The nose temperature (T_n) and the time (t_n) up to the nose point in the C.C.T. curves were 1018 K and 0.93 s respectively. The critical cooling rate for glass formation defined by (T_m−T_n)⁄t_n is evaluated to be 110 K/s. Further, the time interval between Cs and Ct is as short as 0.2 s and the fast growth reaction is attributed to the easy formation of the nonequilibrium crystalline phases and the increase in temperature caused by the precipitation-induced recalescence.

View full abstract

The glassy alloy containing 2 at% Ge in the Fe_74−xAl₅P₁₁C₆B₄Ge_x system was found to cause the extension of the supercooled liquid region, ΔT_x(=T_x−T_g) defined by the difference between crystallization temperature (T_x) and glass transition temperature (T_g) to 60 K, though the ΔT_x value of the Ge-free Fe₇₄Al₅P₁₁C₆B₄ glassy alloy is 45 K. The remarkable increase in ΔT_x is attributed to the increase in T_x exceeding the degree of the increase in T_g. The Fe₇₂Al₅P₁₁C₆B₄Ge₂ glassy alloy has good bending ductility and ferromagnetism with Curie temperature at 590 K, in addition to the appearance of the wide supercooled liquid region. The crystallization of the glassy alloy containing 2%Ge takes place through a single stage and the crystallized structure consists of α-Fe, Fe₃P, Fe₃C, Fe₃B and Fe₂B phases. There are no appreciable changes with Ge content in the crystallized structure and the single-stage crystallization process leading to the simultaneous precipitation of the five crystalline phases. The effectiveness of 2%Ge on the increase in ΔT_x is presumably due to the retardation of crystallization caused by the decrease in the formation tendency of Fe₃C, Fe₃B and Fe₂B phases resulting from the existence of Ge which is soluble to Fe and insoluble to B and C.

View full abstract