One of the authors (T. Yokobori) has made an investigation of the fatigue microcrack initiation and propagation leading to final fracture in plain low carbon steel and in low carbon tempered martensitic high strength steel using the plastic replication method and a scanning electron microscope. As a part of this research program, the present investigation was undertaken to make clear the nature of fatigue crack in high strength eutectoid steel.
It is found that almost all the fatigue microcracks are initiated from the aluminum oxide type inclusion. The initiation of these microcracks occurs at only 5–20% of the total life and they propagate in such a process that the microcrack initiated from a high stress concentrating inclusion joins another one originated from other inclusions, in which case the former grows predominantly to considerable size and finally leads to eventual fracture. The part through fatigue crack grows as a semicircular configuration with the inclusion as the center.
The fatigue crack propagation rate obeys the power relation, dc⁄dN=A(ΔK)^δ, where δ=1.8. Comparing this result with those of plain low carbon steel and low carbon tempered martensitic high strength steel, the higher the strength level, the lower the fatigue crack propagation rate. The value of δ decreases with increasing strength level. The trend is in good accord with the kinetic theory of fatigue crack propagation proposed by Yokobori.

View full abstract

The influence of deformation twinning on the alloy type texture formation in fcc metals was examined by rolling of copper and Cu-6 at%Al alloy single crystals having near (123)[\bar2\bar11] and (111)[0\bar11] orientations.
Copper crystals exhibited deformation only by slip and developed the {123}⟨211⟩ texture. In the alloy crystals, deformation twinning occurred extensively from low rolling reductions. The twinned crystals rotated not toward {110}⟨001⟩ but toward {111}⟨101⟩ and then a double {111}⟨101⟩ intermediate texture was formed.
The displacement from {123}⟨211⟩ to {111}⟨101⟩ and the stabilization of {111}⟨101⟩ in the alloy crystals were discussed in terms of twinning shear rotation and latent hardening. Formation of the components having a (111) plane parallel to the rolling plane was concluded to be one of the important characteristics of the alloy type rolling texture.

View full abstract

The change in resistivity during extension at 77 K was measured on a Cu-4 wt%Ti alloy aged at relatively low temperatures, 150–450°C. The resistivity of each aged specimen increased with straining at 77 K. The resistivity increment for a strain of 5% became maximum for the specimen aged for a certain period of time at each temperature. The aging time giving the maximum increase in resistivity of the specimen, t_max, tended to be shorter as the aging temperature increased. It was shown that the resistivity increase observed was due largely to the destruction of the composition modulation or the modulated structure present in the aged specimen. It was suggested that the time t_max indicated a characteristic time t_s at which spinodal decomposition of the alloy was completed and the amplitude of the composition modulation became maximum at a given temperature. t_max coincided well with t_s which was determined from X-ray small-angle scattering experiments on this alloy. Based on Cahn’s theory of spinodal decomposition, the temperature dependence of t_s was found to be given by the relation:
(Remark: Graphics omitted.)
where T_s was the coherent spinodal temperature and E the activation energy for diffusion of solute atoms in the alloy. With the observed values t_s\simeqt_max and T_s\simeq350°C, we obtained E\simeq0.8 eV . This value was reasonable for the diffusion of Ti atoms in the alloy during the early stages of aging at relatively low temperatures.

View full abstract

Relation between the mechanical properties and interfacial reaction of composites containing tungsten or molybdenum fibers in the copper alloy matrix was investigated. The composites were prepared by infiltration and hot pressing methods. The thickness of the reaction zone in composites which was detected from electron microprobe examination was regulated by changing the annealing time at 850°C. The strength and Young’s modulus of composites were measured by an Instron type tensile machine at room temperature.
The results obtained were as follows:
(1) The mechanical properties of composites with Cu–Ni and Cu–Mn alloy matrices prepared by the infiltration method were damaged by the products of fiber-matrix reaction and recrystallization of fibers.
(2) Nickel and manganese have been added to the copper matrix so that the reaction zone was formed at the fiber-matrix interface. The thickness of the reaction zone in hot pressed composites was regulated quite easily by changing the annealing time.
(3) An appropriate thickness (5∼10 μ) of the reaction zone at the interface contributed to the improvement of the strength of composites. The thickness of the reaction zone increased further, the strength decreased rapidly.
(4) A positive contribution of the reaction zone to the strength of the composites whose orientation lay on 15° and 30° was also observed.

View full abstract

The activation energy for transient creep by a differential test has been measured as follows. The temperature of the specimen is raised to the testing temperature in advance by a furnace placed outside a vacuum vessel. The heat necessary for raising the temperature of the specimen is instantly given by applying half a cycle of a very large alternating current (1400 A) of 50 Hz to the specimen, and a small current is applied so as to keep the temperature constant. This procedure makes it possible to raise the temperature of specimens with a cross section of 2×10 mm² by about 5°C for 10 ms. The activation energy for transient creep of alpha iron obtained at 700°C shows a higher value than that for self-diffusion when the creep strain is nearly zero and decreases rapidly to a minimum value (first region). Then the value increases to a peak value (second region) and decreases again gradually to that for steady-state creep (third region) as the creep strain increases. The first and second regions are observed below the creep strain of about 3%. The phenomenon is also observed in Fe–19Cr alloy. Accordingly, it is thought that the rate controlling process of the transient creep is not such a simple one as those proposed so far but a complex one which is composed of several stages.

View full abstract

The transient creep characteristics of polycrystalline alpha iron were investigated mainly at temperatures ranging from 500 to 650°C using the constant stress creep equipment with a servo divider and the rapid heating method developed newly for determining the activation energy for transient creep. The results obtained are summarized as follows: (1) The transient creep strain exhibits a normal time dependence and inverted transient creep has not been observed. The stress dependence of the initial creep rate, V_i, which is the rate just after applying a load is larger than that of the steady-state creep rate, V_s, and the value of V_i⁄V_s, increases as the stress increases within the range up to about twice the 0.01% offset stress. (2) Of some proposed equations for creep time relation, an equation based on the assumption given by Akulov satisfies the experimental results fairly well. The calculated values of the creep rate approximate the experimental ones at the creep strain more than 1%. (3) The activation energy for transient creep depends on the creep strain. The dependence can be divided into three regions at 700°C as reported previously. However, the first region is not observed at 800°C, and the values of the activation energy obtained at the creep strain of nearly zero are smaller than that of self-diffusion on the contrary to the results at 700°C.

View full abstract

Microduplex grain structures formed on recrystallization in a two-phase (α+γ) temperature range in cold rolled martensitic Fe-16%Ni alloys have been studied in regard to the influence of the third element, Nb or Ti. The microduplex structures in the binary iron nickel alloy consist of globular γ grains smaller than 2 μm and surrounding α grains having similar sizes and shapes. With the addition of 0.17%Nb microduplex grain sizes are reduced to a quarter of those in the undoped alloy. The grain refinement effect due to Nb or Ti is most striking when the deformed martensite is previously aged at 410°C before recrystallization. It is suggested that the third elements forming intermetallic compounds with Ni play a dominant part in the growth inhibition of the recrystallized grains in the deformed martensite matrix.
Microduplex grains originating from undeformed martensite are acicular and their sizes are almost invariable with the third elements.

View full abstract