The effect of hydrostatic pressures on thermoelastic martensitic transformations in aged Ti-51at%Ni and ausaged Fe-31.9Ni-9.8Co-4.1Ti (at%) shape memory alloys has been investigated by electrical resistivity measurements under hydrostatic pressures up to 1.5 GPa. The results obtained were as follows: Transformation temperatures of the B2\ ightleftarrowsR transformations (defined as M_s(R) and A_s(R) in the present study) for two kinds of Ti–Ni alloys aged for short and long periods were independent of pressure. M_s and A_s temperatures of the successive R\ ightleftarrowsB19′ transformations were linearly increased with increasing pressure for the short aged Ti–Ni alloy, but remained constant for the long aged one. On the other hand, the M_s and A_f temperatures in the ausaged Fe–Ni–Co–Ti alloy were inversely decreased with increasing pressure. A thermodynamic analysis on the pressure dependence of the thermodynamic equilibrium temperature, which was obtained from the measured transformation temperatures, showed that the effect of hydrostatic pressures on the B2\ ightleftarrowsR and R\ ightleftarrowsB19′ martensitic transformations in the short aged Ti–Ni alloy was successfully explained by a modified Patel and Cohen’s equation derived in the previous study, but was not for the γ\ ightleftarrowsα′ transformation in the ausaged Fe–Ni–Co–Ti alloy. The disagreement in the ausaged Fe–Ni–Co–Ti alloy is considered to be caused by the invar nature of the alloy, as in the Fe–Ni and Fe–Ni–C invar alloys previously studied.

View full abstract

Pure iron was ion-nitrided in a mixed gas (80 vol% N₂ + 20 vol% H₂) under a reduced pressure of 665 Pa at 803, 923 and 1013 K and then quenched in water for the elucidation of the ion-nitriding behavior accompanied with phase transformations above the eutectoid transformation temperature of 863 K for the Fe–N system.
When nitrided at 803 K, the nitrided specimen surface consists of two layers as usual nitrided cases do; (i) a compound layer with a hardness of about Hv 700 which is mainly composed of γ′-Fe₄N, and (ii) a diffusion layer of about Hv 250.
When nitrided at 923 K, the nitrided specimen surface consists of; (i) a compound layer of about Hv 750, mainly composed of γ′-Fe₄N, under which are (ii) an inner layer of γ austenite of about Hv 400 and (iii) a diffusion layer.
When nitrided at 1013 K, the nitrided specimen surface consists of; (i) a thin compound layer, (ii) an inner hardened layer of γ and α′ martensite, of which the hardness increases from Hv 400 to Hv 700 toward the inside corresponding to the content of the α′ phase, and (iii) a diffusion layer.
To explain the formation of the relatively thick hardened layers of austenite at 923 K and martensite at 1013 K, the nitrogen concentration at each temperature and the degree of the stability of the γ phase when quenched in water after nitriding were examined.

View full abstract

The formation ranges of amorphous and crystalline hydrides by hydrogenation of C15 Laves compounds RNi₂ (R=Sm, Gd, Dy, Ho, Er) were examined by X-ray diffractometry (XRD) as a function of both the hydrogenation temperature and the hydrogen pressure. An amorphous single phase was formed below about 500 K independent of the hydrogen pressure for SmNi₂. On the other hand, it was formed in the range of 300–523 K and 2.0–5.0 MPa H₂ for HoNi₂, and in the range of 400–500 K and 4.0–5.0 MPa H₂ for ErNi₂.
The crystallization behavior of the hydrogen-induced amorphous a-RNi₂H_x alloys was investigated by differential scanning calorimetry (DSC) and XRD. The a-RNi₂H_x alloys decompose into c-RNi₅+c-RH₂ for R=Sm and Gd, while they decompose into c-RNi₃+c-RH₂ for R=Dy, Ho and Er. The activation energies for crystallization in these amorphous alloys were almost constant at about 150 kJ/mol in spite of their different crystallization behavior. The enthalpy changes for crystallization were about 3.7 kJ/mol for all the amorphous alloys except for a-ErNi₂H_x. The crystallization temperatures of these amorphous alloys ranged from 628 to 672 K.
The formation process of the hydrogen-induced amorphous a-RNi₂H_x (R=Gd, Dy) alloys was studied by XRD and DSC as a function of the hydrogenation time. Two kinds of the amorphization processes were observed. Amorphization of DyNi₂ progressed by the nucleation and growth of the hydrogen absorbed amorphous phase. On the contrary, amorphization of GdNi₂ progressed by the polymorphous transformation from the hydrogen absorbed Laves compound to the amorphous phase. The Johnson-Mehl-Avrami (JMA) plots for the process of amorphization gave the numerical exponent n=3.9 for DyNi₂ and n=0.5 for GdNi₂, respectively. The difference in the process of hydrogen-induced amorphization in these compounds was discussed.

View full abstract

For rapidly solidified Al–Cu–Fe quasicrystals, detailed metallographic investigation on the stability of the quasicrystalline phase has been carried out. It has been shown that the quasicrystalline phase itself shows distinct changes in morphological features. Additionally the quasicrystalline phase is not entirely stable at 873 K, as has hitherto been believed. It transforms partially to a crystalline phase accompanied by repartitioning of solutes. The studies of repartitioning of the solute elements and diffraction of the transformed phases show that the product phase is the same as the coexisting crystalline phase found in small quantities in the as spun ribbons. An examination of the faceted interfaces between the quasicrystalline and crystalline phases suggest a transformation towards equilibrium state. The observed changes are related to the defect state in the quasicrystalline phase.

View full abstract

The crystal structure of the ζ₂′ Au-49.5 at%Cd martensite, which has been controversial for 50 years, has been determined by using single crystal data collected by four-circle diffractometer with Mo Kα radiation. The space group is P3, trigonal, which has no center of symmetry. This result is in disagreement with that (P\bar31m) reported by Vatanayon and Hehemann. The lattice constants are a=0.8095(3) and c=0.57940(6) nm and 18 atoms are present in the unit cell. The structure was refined by the full-matrix least-squares method to a final R factor of 7.8% (wR=4.1%). The transformation mechanism is also discussed and it is shown that the ζ₂′ martensite is created by three ⟨011⟩⟨0\bar11⟩ transverse displacement waves introduced simultaneously. Simulations with displacement waves were also made and a satisfactory results was obtained.

View full abstract

Experimental results on superplasticity in a titanium aluminide, Ti₃Al, have been reported. Elongations of over 500% can be obtained in this intermetallic alloy. Preliminary results on microstructural evolution of the Ti₃Al alloy show that dynamic grain grown occurs and the volume fraction of α₂ phase decreases during superplastic deformation. The superplastic properties of Ti₃Al are compared with those of another aluminide, Ni₃Al, in terms of the effects of testing temperature and strain rate. Possible mechanisms of superplastic deformation of these intermetallic alloys are also proposed.

View full abstract

In order to clarify the mechanism of change in molten pool geometry by plasma arc heating, the effects of oxygen contents of Fe–O melts on the geometry and flow behavior of molten pool were investigated. Also, the temperature coefficients of surface tension of Fe–O melts were examined in a temperature range from 1823 to 2023 K.
The depth of molten pool increased and the width of it decreased with increasing oxygen contents. Outward flow was observed at the surface of molten pool in the case of a low oxygen level and inward flow was observed in the case of a high oxygen level. The temeprature coefficient of surface tension of Fe–O melts changed from a negative to positive value with increasing oxygen content. The ratio of depth to width of molten pool increased linearly with increasing temperature coefficient of surface tension. The geometry and flow behavior of molten pool were discussed in relation to the change of the temperature coefficient of surface tension with oxygen.
It was concluded from the experimental results that the Marangoni flow induced by the temperature difference in molten pool predominantly affect the geometry of molten pool.

View full abstract

Using a zirconia solid electrolyte and separate gas chambers around the electrodes, a reproducible EMF as a function of temperature was obtained with following cell.
−Pt|Cu, Cu₂O|ZrO₂-electrolyte|Ag–Cu, Cu₂O|Pt+
The EMF for several compositions of Ag–Cu alloys in the solid solution was measured. The thermodynamic activities of both components as well as the partial and integral molar quantities were determined. All activities showed a large positive deviation from Raoult’s law, this implies a demixing tendency in this system. The maximum solubility of Cu in solid Ag at 1053 K was determined as 13.0(±0.5) mol% by extrapolation.

View full abstract

The effect of CaO additions on selenium distribution between sodium-borate slag and copper metal phases was studied at 1473 K under reducing conditions. It was noted that with increase in CaO content (10 to 50 mass%) in slag, the Se distribution coefficient increases. It also increases with increase in initial Se concentration in Cu (0.35 to 0.45 mass%). The results show that the Se distribution coefficient ‘L’ increased by about two times (3 to 5.9) on addition of 50 mass%CaO to sodium-borate slag (Na₂O mass%/B₂O₃ mass%=0.86) for Cu-0.45 mass% Se alloy.

View full abstract

This study discusses TLP bonding of Ni-base superalloys Mar-M247 and IN939, using filler metals specially designed and fabricated into flexible foils by rapid solidification processes. Filler metals, selected by metallographic studies, produce almost the same microstructure in TLP bonds as in the base metal.
The mechanical properties for TLP bonds have been investigated by stress rupture tests, which are the severest tests used to evaluate bonding strength.
The results have also revealed that sufficient bonding pressure was indispensable for Mar-M247 to obtain high bonding efficiency. The TLP bonding efficiency for IN939 was obtained easier than Mar-M247.

View full abstract

A method of producing aluminum profile rods by the moldless upward continuous casting process using formers was investigated. Effects of the configuration of the former, the superheat of molten metal, and the upward withdrawal speed on the ratio of cast strand formation (Ratio of the diameter of circumscribed circle of cast strand to that of the nozzle of the former), the macrostructure, and the surface conditions of the cast strand obtained were examined. Aluminum profile rods having various cross-sectional configurations and also longitudinal tapers could be produced by using formers made of refractory materials which determined the configuration of molten metal pillars formed on withdrawing upwardly. The former should be a “hot” one classified by Kostiegov et al. and its nozzle position should be lower than the molten metal surface by 5∼10 mm. The cast strand was composed of several columnar grains solidified completely unidirectionally along its axes. The ratio of strand formation was determined by the size of the nozzle, the superheat of the molten metal, the withdrawal speed, and the strand cooling intensity. The profile rods with no taper could be produced most easily when the ratio of strand formation was kept below 100%.

View full abstract

In order to investigate the differences in the properties of the coated cemented tungsten carbides, WC/Co, by the PVD and CVD processes, indentations impressed by the Vickers hardness tester on WC/Co coated with TiC films were observed with an acoustic microscope. The difference in debonding of the coated surface films from substrate materials and cracks originating from the Vickers indentations could be analyzed by acoustic microscopy. Observation made by the microscope operated at frequencies, 200 and 400 MHz, revealed the vertical cracks under the films coated by the CVD process, and the debonding of the coating film from the substrate could be observed for the PVD coated films at much more higher frequencies, 600 and 800 MHz. The diffence in the surface wave velocities obtained from V(z) curves for the PVD and CVD coated specimens suggested the existence of higher residual stress in films produced by the PVD process. The observed images were discussed by comparing with the hardness test data and SEM fractography.

View full abstract

A nanoscale bcc film with a grain size of 10 to 20 nm was found to be prepared by annealing sputtered Fe–Zr–B amorphous films with solute contents of 4 to 9 at% Zr and 1 to 6 at% B for 3.6 ks in the temperature range of 823 to 923 K. The bcc Fe_89.6Zr_7.7B_2.7 film obtained by annealing at 923 K exhibits a high magnetization (B₈₀₀) of 1.64 T at an applied field of 800 kA/m and a high permeability (μ_e) of 3200 at 1 MHz. It is notable that the B₈₀₀ and μ_e values are comparable to the highest values for various kinds of soft magnetic films reported so far. The reason for the appearance for the good soft magnetic properties is presumably due to the decrease in apparent magnetic anisotropy by the formation of the nanocrystalline bcc structure and the reduction of magnetostriction caused by the dissolution of Zr into the bcc phase.

View full abstract