The crystallization processes of roller-quenched amorphous Pd_xNi_ySi_z [(x, y, z)=(70, 10, 20), (60, 20, 20), (50, 30, 20), (55, 30, 15)] alloys during isothermal aging are reported. The process was investigated with thermal analysis, electrical resistance measurement and X-ray analysis, and T-T-T diagrams of these alloys were obtained.
The crystalline Pd₃Si and Pd-rich silicide were formed in each alloy. These crystallization, however, are greatly influenced by Ni atoms, that is, the onset time of crystallization retarded with increasing Ni concentration.
In Ni-rich alloys (≥30 at%), a number of Ni-based crystallites are formed before crystallization of Pd₃Si and Pd-rich silicide, and these crystallites increased the electrical resistance of specimens.
X-ray analysis of the crystallized sample showed that the stable Ni-based compound such as Ni₃Si was not formed though Pd₃Si was not detected in each alloy, and this phenomenon suggested that the Pd–Si bonding was stronger than the Ni–Si bonding.

View full abstract

The activity coefficient of oxygen at infinite dilution in liquid copper at 1573 K and the oxygen potential over Cu(l)+Cu₂O(s, l) have been measured using the following galvanic cells:
(I) Pt/Ni(s), NiO(s)/ZrO₂+CaO/O(in liquid Cu)/LaCrO₃·Pt,
(II) Pt/Ni(s), NiO(s)/ZrO₂+CaO/Cu(l), Cu₂O(s, l)/LaCrO₃·Pt.
The results are given by
γ₀°=0.287±0.005 at 1573 K,
Δμ(O₂)⁄J mol⁻¹±300=−(400000±1500)+(191.36±0.98)(T⁄K) at 1355∼1493 K, and
Δμ(O₂)⁄J mol⁻¹±400=−(256200±6000)+(95.18±3.91)(T⁄K) at 1493∼1560 K,
where the errors are given as standard deviations.
From the results described above and those for cell (I) by other investigators, the standard free energy change ΔG°(0) for the reaction
1/2 O₂ (g, 101.325 kPa)→O (in liquid Cu, N₀=0.01)
and the first order self-interaction coefficient of oxygen in liquid copper ε₀⁰ are determined as follows:
ΔG°(0)⁄J mol⁻¹±600=−(82400±2900)+(4.00±1.98)(T⁄K) at 1373∼1573 K, and
ε₀⁰±0.8=−3440⁄(T⁄K)−4.47 at 1373∼1573 K, respectively.
Furthermore, using the value of ε₀⁰ and the Cu–Cu₂O phase diagram, the chemical potentials of Cu(l) and Cu₂O(l) for Cu(l)+Cu₂O(s, l), which are respectively relative to pure Cu(l) and stoichiometric Cu₂O(l), are evaluated as
(RTlna_Cu)⁄J mol⁻¹±10=−73686+454.85(T⁄K)−55.554(T⁄K)ln(T⁄K) at 1358∼1483 K,
(RTlna_Cu)⁄J mol⁻¹±10=2760−2.41(T⁄K) at 1508∼1583 K, and
(RTlna_Cu2O)⁄J mol⁻¹±20=7130−5.64(T⁄K) at 1508∼1583 K.

View full abstract

The oxygen potentials for Ni(s, l)+NiO(s)+NiAl₂O₄(s) and Ni(s, l)+NiO(s)+NiGa₂O₄(s) have been measured using solid oxide galvanic cells of the type:
Ni(s, l), NiO(s), NiX₂O₄(s)/ZrO₂+CaO/air: (X=Al, Ga).
The results are expressed as follows:
(i) for Ni(s, l)+NiO(s)+NiAl₂O₄(s),
Δμ(O₂)⁄J mol⁻¹±300=−(467500±2200)+(168.6±0.7)(T⁄K) at 1285 to 1558 K,
Δμ(O₂)⁄J mol⁻¹±400=−(456800±4900)+(161.9±2.7)(T⁄K) at 1726 to 1835 K,
(ii) for Ni(s, l)+NiO(s)+NiGa₂O₄(s),
Δμ(O₂)⁄J mol⁻¹±200=−(469200±700)+(170.0±0.5)(T⁄K) at 1271 to 1528 K,
Δμ(O₂)⁄J mol⁻¹±500=−(460300±5500)+(163.8±3.1)(T⁄K) at 1717 to 1853 K,
where the errors are given as standard deviations.
The solubilities of oxygen in liquid nickel equilibrated with NiO(s) and NiAl₂O₄(s), and with NiO(s) and NiGa₂O₄(s) have been measured by a phase equilibration technique. They are given by
(i) for liquid nickel in equilibrium with NiO(s) and NiAl₂O₄(s),
lnN_o±0.012=−(24070±300)⁄(T⁄K)+(9.346±0.167) at 1720 to 1850 K,
(ii) for liquid nickel in equilibrium with NiO(s) and NiGa₂O₄(s),
lnN_o±0.016=−(23870±470)⁄(T⁄K)+(9.215±0.262) at 1725 to 1845 K.
From this work, it was found that
(1) the solubilities of Al₂O₃ and Ga₂O₃ in NiO were negligible up to about 1573 and 1553 K, respectively,
(2) the oxygen activity and the oxygen concentration in liquid nickel coexisting with NiO(s) and NiAl₂O₄(s), and with NiO(s) and NiGa₂O₄(s) satisfied the relation between those in the Ni–O system.

View full abstract

Interface segregation behavior in α(fcc)⁄β(bcc) brass two-phase bicrystals and polycrystals was investigated by means of EPMA. The following facts were made clear from the present work.
(1) Segregation profile in polycrystals was recognized as both the enrichment of Zn in β phase and the depletion of Zn in the α phase adjoining to the α⁄β interface. In two-phase bicrystals, however, only the enrichment of Zn in the β phase adjoining to the α⁄β interface was observed.
(2) The width and height of segregation in two-phase bicrystals increased with increasing annealing temperature. More pronounced segregation was observed in the bicrystals than in the polycrystals.
(3) It was explained by the crystallographic considerations that the segregation profile observed was due to the character of the plane-matched interface structure. Furthermore, the interfacial energy and the lattice strain energy before and after the segregation were estimated. The energetic consideration also explained the experimental results reasonably.
(4) A lack of the depletion of Zn in α phase adjoining to the interface in two-phase bicrystals was considered to result from the higher mobility of the interface than that of Zn atoms in the α phase.

View full abstract

A non-destructive in-depth analysis of a single phase brass has been done by a combination of X-ray induced AES, EPMA and X-ray diffraction. The analysing depth by this combination covers the depth range from 1 nm to 10 μm. The anlytical composition \barC(L) by a method with analysing depth L is converted to the true composition C(L) at depth L by using a certain composition-distribution model and the following relation;
\barC(L)=∫₀^∞C(x)Y(x,L)dx⁄∫₀^∞Y(x,L)dx
where Y(x,L)=exp{xln(1−G_L)⁄L} for XAES and X-ray diffraction and =φ(ρx,L)·exp(−χρx) for EPMA, φ(ρx) is the ionization distribution function and G_L is a constant. Samples were revealed to have a large dezincification zone at the depth less than 1 μm. The largest difference between \barC(L) and C(L) in this case appeared in the depth range between 10⁻² and 10⁻¹ μm.

View full abstract

Creep characteristics of a bcc Fe-4.1 at%Mo solid solution were investigated under a constant stress of 6∼40 MPa at 1100∼1225 K in an argon atmosphere. The stress exponent for the steady-state creep rate was 3.2±0.2 and the apparent activation energy was 320±8 kJ/mol.
Creep curves were of inverse- or sigmoidal-types. After an increase in stress during steady-state creep, an inverse transient creep curve was obtained without instantaneous plastic strain. The average internal stress during the steady-state creep, determined by a stress-transient dip test applying a back-extrapolation technique, was smaller than the applied stress. These results indicate that the role of dislocations which glide viscously should be taken into account in the discussion of the steady-state creep process in this alloy under the present creep condition.
Substructures developed during the steady-state creep were observed by transmission electron microscopy. Sub-boundaries consisting of dislocation networks as well as rather homogeneously-distributed dislocations within subgrains were always observed. The dislocation density within subgrains, the total dislocation density, the length of sub-boundaries per unit area, and the mean spacing of dislocations in sub-boundaries were determined as a power function of stress. The stress exponents of these quantities were 1.5, 1.3, 0.5 and −0.6, respectively.
The creep characteristics are discussed in terms of the motion of dislocations which drag the Cottrell solute atmosphere, on the basis of experimental data on the average effective stress and the dislocation density within subgrains. The values of the stress exponent and the activation energy obtained theoretically are in reasonable agreement with the experimental ones. The calculated value of the steady-state creep rate, however, is about one order of magnitude smaller than the actually observed one.

View full abstract

Niobium and niobium alloys have received considerable attention as engineering materials for heat engine applications. In the past several years, the carbothermic reduction-electron beam melting combination method has made it possible to prepare pure niobium metal. In this work, the possibilities and conditions for winning the pure niobium by this process have been examined with regard to the following items: (1) Working conditions and analytical results on the Nb–C–O solid solutions obtained by the carbothermic reduction of Nb₂O₂ in vacuum. (2) Vacuum degassing in the solid state. (3) Refining of niobium by electron beam melting. (4) Evaporation deoxidation in the form of niobium suboxides.
The mechanism of refining of the niobium metal by the present method can be characterized by the next steps, carbon is removed by degassing as CO and the elimination of oxygen takes place in the form of CO and volatile suboxides such as NbO and NbO₂. Both phenomena seem to occur during the refining period in the solid state under the vacuum at 2200 K and the subsequent electron beam melting. Accordingly, the carbon elimination is possible only when an excess of oxygen is present at the beginning of electron beam melting. On the other hand, even after the decrease of carbon, the remaining oxygen can be removed by the suboxide evaporation.
With this process the high purity niobium in which both carbon and oxygen contents are reduced to a sufficient degree is obtainable by favorable conditions of reduction and melting.

View full abstract

Movement of three-fold nodes of boundaries in high temperature creep has been examined under a 25% stress of the apparent yield stress in polycrystalline aluminum with the following results. (1) Contribution of the boundary migration to the boundary movement is larger than that of the boundary sliding, and increases with increasing creep temperature. (2) Movement of the three-fold nodes is enhanced when the strain marking is induced by the boundary sliding in a grain toward which the boundary migration occurs. On the contrary, the movement is retarded when the strain marking is induced in the opposite grain. (3) The results have also been examined by in situ creep deformation with a high voltage electron microscope, and the relation between the boundary movement and the dislocation density in the composite grains has been clarified.
Based on the results, it is concluded that the migration of boundaries including the three-fold nodes in high temperature creep depends on the difference in the density of dislocations terminated on each individual boundary among composite grains.

View full abstract