Amorphous alloys exhibiting a wide supercooled liquid region above 100 K were found to form in a compositional range from 0 to 3%Co, 0 to 15%Ni and 10 to 23%Cu in Zr₆₅Al_7.5Cu_2.5(Co_1−x−yNi_xCu_y)₂₅ system by melt spinning. The temperature span ΔT_x(=T_x−T_g) between glass transition temperature (T_g) and crystallization temperature (T_x) reaches as large as 127 K for Zr₆₅Al_7.5Ni₁₀Cu_17.5. The T_g and hardness (H_v) increase from 622 to 685 K and 426 to 502 with increasing Co content while the T_x decreases from 749 to 690 K, resulting in the decrease of ΔT_x from 127 to 30 K with increasing Co content. The compositional effect on T_g, T_x, ΔT_x and H_v indicates that there is no close relation between the magnitude of the attractive bonding force and the stability of the supercooled liquid. The high stability of the supercooled liquid against the nucleation and growth of a crystalline phase in the limited composition range seems to result from a highly dense random packing structure consisting of atoms with an optimum atomic size ratio and a large negative enthalpy of mixing.

View full abstract

The present study by X-ray and neutron diffraction methods is concerned with the ω-like peak appearing at the positions (H±2⁄3, K±2⁄3, L±4⁄3) and (H±2⁄3, K±2⁄3, L\mp4⁄3) in the reciprocal lattice of the quenched β₁ phase with the Heusler type structure of AuCuZn₂. The obtained results indicate evidently that the ω-like peak is neither due to the ω phase formation as is the case of Zr alloy nor precursor phenomenon of the martensitic transformation. It is originated from a certain kind of regular arrangement of vacancies in the incompletely ordered B2 phase: the vacancies are responsible for the ordering reaction to the Heusler type structure occurring at about 670 K. They are frozen into the Heusler phase formed by quenching, and hence the quenched Heusler phase exhibits the ω-like peak as observed. The peak disappears with vacancy annihilation from the Heusler type β₁ phase. Characteristics of both the [111]LA and [11\bar2]TA phonon dispersion relations also support this conclusion.

View full abstract

The heat treatment under a low partial pressure oxygen atmosphere has previously been developed to get an excellent oxidation resistance of intermetallic compound TiAl in the cyclic oxidation test at 900°C. In this study, oxidation behaviors of Ti-31∼39 mass% Al alloys and Ti-36 mass% Al alloys with different oxygen contents up to 0.38 mass% were investigated in cyclic oxidation tests at 900 and 950°C from a viewpoint of the effect of the heat treatment stated above. The mass gain due to cyclic oxidation test of the specimens without the heat treatment decreased with increasing the aluminum content of the alloy, while it was independent of the oxygen content at 900 and 950°C. The specimens with higher oxygen contents than 0.12 mass%, however, were found to form oxide TiO₂ even at such a low temperature as 700°C. The effect of the heat treatment was reduced at 950°C tests. The higher the aluminum content of the alloy, the longer the duration in which the excellent oxidation resistance was maintained in the cyclic test. The heat treatment at 1000°C improved the oxidation resistance of the alloys containing oxygen up to 0.12 mass%, and was not effective for higher oxygen alloys. The optimum condition of the heat treatment under a low partial pressure oxygen atmosphere was concluded to be a vicinity of a temperature of 975°C and an air pressure of 1.3×10⁻³ Pa for 16 h.

View full abstract

Translucent alumina ceramics with extremely high in-line optical transmittance to visible light and high mechanical strength were produced by applying HIP process to a sample which was densified at a relatively low temperature after injection molding. Measurements of bulk density and average grain size, X-ray diffraction (XRD) and secondary ion mass spectroscopy (SIMS) analysis, and SEM and TEM observations were conducted. The bulk density was higher than 3.985 kg/m³, average grain size was between 0.82 and 1.43 μm, and the lattice constants were a=0.476 nm and c=1.300 nm. The segregation of impurities at grain boundaries and the grain boundary interface was not observed. The in-line transmittance increased with the reduction of the average grain size of the translucent alumina ceramics. It was confirmed that in the translucent alumina ceramics, optical scattering does not occur due to the increase in the number of grain boundaries and pores, segregation of impurities and distortion of lattices as conventionally reported, and hence does not conform to Fresnel’s law. Based on these findings, it was concluded that the improvement of optical transmittance can be realized by reducing the crystal grains to smaller than 1 μm as well as by the conventional method of growing them to larger than 20–30 μm.

View full abstract

The anomalous X-ray scattering (AXS) technique is applied to obtain the environmental radial distribution functions (RDFs) around germanium atoms in glassy (1−x)GeO₂·xP₂O₅ (x=0.10, 0.18 and 0.28), from the intensities measured at the Ge K absorption edge using synchrotron radiation. The analysis of the shape of the RDFs shows that the distance of Ge–O pairs increases with addition of P₂O₅ and the number of coordinated oxygen atoms around germanium atoms also increases from four with the addition.

View full abstract

The ternary chamical diffusion has been studied in the iron-nickel-chromium system in the gamma phase at the temperature of 1223 K. The concentration penetration profiles of the diffusing elements have been established employing an electron probe microanalyser. The major and cross diffusion coefficients have been evaluated using the Boltzmann-Matano method as extended to the ternary system. The major diffusion coefficients are positive, while the cross diffusion coefficients are negative. The concentration denpendence of diffusion coefficients has been rationalised and can be expressed as the function of concentrations of nickel and chromium. The interaction parameters have been estimated from the variation of ratio of the major to cross diffusion coefficients with concentration. The estimated interaction parameters indicate that the diffusive interaction between nickel and iron is much stronger than that between chromium and nickel or chromium and iron. Tracer diffusion coefficients have been estimated from the ternary diffusion coefficients and are in the order D_Cr^*>D_Fe^*>D_Ni^*. Intrinsic diffusivities have also been evaluated.

View full abstract

A new method has been proposed for the estimation of the three-dimensional grain size distribution from the two-dimensional distribution measured on the cross section of polycrystalline material. In this method, twelve types of polyhedra were employed as the grain models. The distributions of the cross-sectional diameters of the individual polyhedra were expressed as probability density functions. On the basis of the functions for each polyhedron, the supposed grain size distribution on the cross section of the material was calculated, and it was compared with the measured one. The comparison was repeated until the agreement between the both distributions. By operating this two-dimensional distribution reversely, the three-dimensional grain size distribution was estimated.
The distribution of the vertex number of polygon-shaped grains on the cross section and that of the face number of polyhedron-shaped grains in three dimensions were calculated from the obtained results and were compared with the measured distributions. There was a good agreement between these distributions.

View full abstract

In order to make clear the phase equilibrium in the Y–Ba–O system, the samples were prepared using the raw materials of Y₂O₃ and BaCO₃ or BaO₂ in natural air or in mixed gas of 21%O₂+Ar, respectively. The phase relation was investigated using X-ray diffraction and EPMA methods. The carbon content was measured by means of oxygen combustion infrared absorption method. From the experimental results, the isobaric sections of the Y–Ba–O system were evaluated. Four compounds were found to exist in the air when using BaCO₃ or BaO₂ as one of the raw materials and also when using BaCO₃ in a mixed gas of 21%O₂+Ar. However, two of them, Y₂Ba₃O₆·CO₂ and Y₂Ba₄O₇·CO₂, were oxycarbonates. On the other hand, by sintering the raw materials of BaO₂ and Y₂O₃ in the mixed gas of 21%O₂+Ar, only two pure oxides, Y₂BaO₄ and Y₄Ba₃O₉, were detected.

View full abstract

Hot dip aluminized layers on mild steel have been studied by optical microscopy, microhardness measurement, X-ray diffraction and electron diffraction. The investigation establishes the presence of an outermost, very soft, Al bearing layer followed by a soft intermetallic layer containing Fe₃Al and a hard intermetallic region consisting of FeAl₃ and Fe₂Al₅. The presence of an outer Al layer is responsible for the material’s low degree of wear resistance and low stability at high temperatures. The possibility of eliminating or minimizing this layer by diffusion annealing was studied at 773, 873 and 923 K, for 10 min–10 h. Aluminum is found to diffuse, eliminating the pure Al layer and giving rise to FeAl₃ initially. Subsequently the phase Fe₂Al₅ forms along with FeAl₃ and finally only Fe₂Al₅ forms. However, this heat treatment is not sufficient to precipitate any of the favourable phases such as FeAl, Fe₃Al which satisfy Pilling-Bedworth’s ratio for providing good compatibility with the matrix. The intermetallic layers are adherent and can endure thermal shock and bending.

View full abstract

Creep tests were carried out on an aluminum composite reinforced with continuous alumina fibers at temperatures of 573 to 773 K to examine the mechanism of high temperature deformation and fracture of the composite.
The creep behaviour was different in the low and high stress ranges. The creep curves in the low stress range showed only mostly the primary or slight secondary stage without reaching the tertiary stage. The creep strain consisted of an irreversible component, possibly due to the plastic deformation of matrix around the misaligned fibers and the fracture of fibers at weak points, beside a reversible component due to the elastic and anelastic deformation of the matrix and fibers, although the former component was very small at lower stresses.
In the high stress range, the creep deformation proceeded by the local fracture of fibers and the plastic deformation of the matrix resulting in fracture of the specimen. The creep curves showed the primary, secondary and tertiary stages.
The critical stress which distinguished between the low and high stress ranges was about 500 MPa at 573 K and 450 MPa at 773 K. The critical stress was consistent with the threshold stress which was recognized in the stress dependence of the minimum creep rate. The critical stress, or the threshold stress is considered to be the minimum stress necessary for the individual cracks of fibers to propagate extensively and to cause the fracture of the whole specimen.

View full abstract

The valence of Cu ions in Cu-compounds is estimated by the energy-dispersive EPMA method. The intensity ratio of the X-ray characteristic lines L_α to K_α for Cu, L_α⁄K_α, decreases exponentially with increasing Cu valence as examined for standard samples: metallic Cu (Cu⁰⁺), Cu₂O (Cu¹⁺), CuO (Cu²⁺) and NaCuO₂ (Cu³⁺). The relationship between L_α⁄K_α and the Cu valence appears as different functions for different accelerating voltages, but is reduced to a universal function after ZAF correction and normalization were made. In addition to the standard samples, NaCuO (Cu¹⁺) and CuF₂ (Cu²⁺) are measured to investigate how the same function is valid.
The valences of Cu ions in a high-T_c oxide superconductor Ba₂NdCu₃O_7−δ are estimated by the above-mentioned method. To avoid large errors introduced from inaccuracy of the ZAF correction for materials with elements having a large mass number, we made an experimental determination of the ZAF correction-coefficient for Ba₂NdCu₃O_7−δ, using the mixture of BaCO₃, Nd₂O₃ and Cu₂O (Cu¹⁺) or CuO (Cu²⁺) of which atomic ratio Ba:Nd:Cu=2:1:3. For Ortho-1 (δ=0.1; T_c=80 K), Ortho-2 (δ=0.3; T_c=30 K) and Tetra (δ=0.6; T_c=0 K, non superconductivity) samples, the Cu valences are estimated at 2.0±0.05, 1.9±0.1 and 1.55±0.2, respectively. For Bi₂Sr₂CaCu₂O_8+δ and Bi_1.8Pb_0.3Sr_1.9Ca₂Cu₃O_10+δ, the Cu valences are estimated by the same method to be 2.0±0.1 and 2.0±0.2, respectively, like in the Ortho-1 Ba₂NdCu₃O_7−δ.

View full abstract

The coefficients for equilibrium partition of solute elements between solid and liquid in freezing aluminium alloys were evaluated thermodynamically, and the availability of the thermodynamic calculation was elucidated from a comparison with the experimental results obtained for the alloys rapid-quenched from a coexisting solid-liquid state and those in the literature. Using the equilibrium partition coefficients, liquidus and eutectic temperatures were calculated for the aluminium-based ternary and quaternary alloys.

View full abstract

This paper examines the effects of Nb and Si on the internal friction, magnetic properties, and rigidity modulus defect of Ni–Co–Nb–Cr alloys either furnace-cooled or water-quenched after heating at 1273 K for 7.2 ks. Both the internal friction and the rigidity modulus are measured with an inverted torsion pendulum, and the magnetic properties by an automatic recording fluxmeter.
As the shear strain amplitude increases, the internal friction grows and reaches a maximum. The maximum internal friction and the shear strain amplitude were defined as Q_max⁻¹ and γ_c. As the Cr concentration increases, Q_max⁻¹ rises to a maximum of 51 to 91×10⁻³, and decreases gradually. For the alloys containing 25 mass% Co and 3 (or 5) mass% Nb, the maximum is obtained at 1 (or 3) mass% Cr with furnace cooling and at 3 mass% Cr (for either Nb concentration) with water quenching. The Q_max⁻¹ values of the alloys containing 45 mass% Co and 3 or 5 mass% Nb are obtained at 3 mass% Cr with furnace cooling and at 1 mass% Cr with water quenching; similarly, those of the alloys containing 65 mass% Co and 3 or 5 mass% Nb are obtained at 3 mass% Cr with furnace cooling and at 1 mass% Cr with water quenching.
Generally, the γ_c value falls as the Cr concentration increases. The variation in coercive force H_c with composition is very similar to γ_c. In the alloys with the highest Q_max⁻¹ value, both H_c and γ_c rise with increasing Co concentration. Both the rigidity modulus defect (ΔG) and Q_max⁻¹ are highest at nearly the same composition, and their variations with composition were also very similar.

View full abstract