It is well known that when steel is hot-quenched to the temperature range of lower bainite formation and is partially transformed isothermally, the residual austenite is stabilized and austenite retained (γ_R) at room temperature is increased. In previous reports, however, the author indicated the existence of the effect of stress on the formation of martensite nuclei in addition to the hitherto known effect of stress on the driving force of martensite transformation, and assumed the coherent embryos of martensite formed due to Bain’s strain to explain the reverse effect of tension and compression upon the transformation. In the present study, the author has investigated how austenite is stabilized when steel is held isothermally in the temperature range of unstable austenite under the load of tension or compression. The experimental results obtained are as follows; (1) When steel samples are held for the incubation time, there is no change in the amount of γ_R irrespective of the holding temperature and the application of stress of positive or negative sign. Therefore, this substantiates the results obtained in the previous report that “the amount of γ_R is dependent upon the temperature range in which the stress is applied”. (2) In the bainite temperature range the incubation time is shortened as a result of the shear stress effect irrespective of its sign. At room temperature, however, tensile stress increases γ_R and compressive stress decreases γ_R (as in case of the stress effect during continuous cooling disclosed in the previous report), owing to the reverse effect of stresses, depending on whether the sign is positive or negative, and on the number of nuclei formed after holding the samples for a certain time. Thus it is confirmed here again that there is an effect of stress on the number of martensite nuclei.

View full abstract

Recent developments in the tantalum refining industry have demanded a handy method of analyzing tantalum ores and related materials such as intermediates and residues, for the control and improvement of the refining processes. By use of a medium quartz spectrograph the authors studied a spectrographic analysis for these materials, in which the sample was fused with a fusion flux, powdered, and then sparked with a Feussner spark source. As a result, a method suitable for routine analysis could be established. The factors investigated in the experiments were the fusion flux, amount of the flux to be added, fusion time, analytical gap, shape of the electrode, particle size and quantity of the sample to be fused, particle size of the fused sample to be sparked, analytical line pairs, and errors due to the sample preparation and the analysis.
The outline of the method obtained was as follows: A mixture of 100 mg of finely ground sample and 50 mg of GeO₂ as an internal standard is carefully fused with 1.5 g of KHSO₄, and the resultant melt is cooled, pulverized to 180 mesh, charged into cup-electrode, and sparked with a graphite counter electrode sharpened to a truncated cone. The standard samples are prepared from synthetic oxides by applying the same fusion treatment as described above. In the various tantalum ores and related materials, Ta, Nb, Fe, Mn, Sn, Ca, Si, Ti, Al, Cr and Mg are determined over the range of 0.05∼100% with a variation coefficient of 3∼5%, and the chemical and spectrographic results are closely consistent with each other.

View full abstract

To investigate the effect of aluminum on the properties of high carbon steels containing 1% carbon and 1.4% chromium, the authors measred the critical point, hardenability, quenched and tempered hardness, retained austenite, hot hardness, dilatation and electron-microscopic structures. The results obtained are as follows: (1) The critical point is raised with the aluminum and silicon contents. The hardness of 65 Rockwell C is obtained by oil-quenching from 840°C for 1% aluminum steel and 870°C for 1% aluminum-1% silicon steel. (2) The rate of softening with tempering temperature decreases in the steels between 200° and 350°C, and the magnitude of this decrease in softening rate is related to the aluminum content. In fact the rate of softening becomes nil at about 300°C in the steels which contain 1% or more aluminum and silicon, and these steels have the hardness of Rockwell C 60 after temering at 350°C. (3) From the results of the dilatometer test, it is found that the temperature at which the third stage contraction can be detected is raised with the aluminum or silicon content. (4) It is estimated that the operating temperature of 1% carbon and 1.4% chromium steel could be raised by 50∼100°C with increase in aluminum content from 0.02 to 1.0%.

View full abstract

This paper deals with changes in the internal friction at room temperature of Hadfield steel by annealing. The specimen was annealed at various temperatures after quenching in water from 1000°C (solution treatment) to which transversal resonant vibrattion was established by the electro-magnetic method. Then, its internal friction was obtained from the free decay of resonant vibration after the driving electric current was switched off.
Experimental results obtained are as follows:
(1) A specimen annealed after solution treatment or as-solution treated was found to have strain amplitude dependent internal friction, but annealing at 500°∼700°C results in somewhat reduced strain amplitude dependence.
(2) A specimen annealed at the temperature range of 500°∼800°C where the precipitation of carbides and the formamation of needle-likes phase (α, ε) takes places, is low in strain amplitude independent internal friction and high in hardness.

View full abstract

Formability of the quenched Ti-2∼8% Al-2∼8% Co alloys was studied by the tensile tests and the cold-rolling test. The values of δ_max⁄σ_(δ=δmax) were calculated as “formability index” from the tensile test data, where δ is the true strain, δ_max is the limit value of the uniform elongation and σ is the true stress. High formability indices were obtained with the specimens transforming from β to martensite by stress. Particularly, the formability index of Ti-4% Al-4% Co alloy which was quenched from 900°C corresponded to commercial pure titanium. Low formability indices were obtained with the specimens having a stable β structure. The specimens having α+Ti₂Co, α+β+Ti₂Co and α+β structures had moderate formability indices. Results of the cold-rolling test indicated that the specimens transforming remarkably to martensite by stress was low in reduction ability, and that the specimens having a stable β phase was high in reduction ability. Particularly, Ti-4% Al-8% Co and Ti-6% Al-6% Co quenched from β phase excelled in reduction ability. This difference may be attributed to the reduction ability during rolling which is not determined by the extent of uniform elongation but by the occurrence of cracks during the deformation beyond the limit of uniform elongation. In other words, owing to the low rate of strain hardening, Ti-4% Al-8% Co alloy quenched from β phase has a small uniform elongation, but it can be deformed heavily beyond the limit of uniform elongation. On the other hand, the specimens forming stress-induced martensite with a high rate of strain hardening have a large uniform elongation, but they show cracking when deformed beyond the limit of uniform elongation. It can be concluded that the specimens having a relatively stable β phase are suitable for the forming which does not require a large uniform elongation, and that the specimens readily forming stress-induced martensite are suitable for the forming in which a large uniform elongation is required.

View full abstract

It has been proved that the interstitially dissoled amounts of N and C can be quantitatively measured from the Snoek peak height by using the transverse vibration method at about 600 c/s. The proportional coefficient between the dissolved amounts of N or C and the Snoek peak height is influenced by grain size and impurities, but the following relations, practically wt% N or C=0.00434×T_n×(Q⁻¹)_max or 0.00430×T_c×(Q⁻¹)_max, are shown, where T_n or T_c are the peak temperature of N or C and (Q⁻¹)_max is the Snoek peak height. It has also been ascertained that the Snoek peak height is inversely propotional to the peak temperature between a hundred and several thousand c/s as predicted theoretically.

View full abstract

The lower yield stress of high purity aluminum-magnesium alloys with Mg contents of 0.86, 2.05 and 3.67% and a commercial alloy of 4.75% Mg, which have been well-known as alloys showing a sharp yield elongation and exhibiting the propagation of Lüders bands during yielding, was investigated under dynamic loading at room temperature and compared with that in the static test with special attention to its grain size dependence, following the Cottrell-Petch equation σ_y=σ_i+kd^−1⁄2. The observed increase of the lower yield stress in the dynamic test from that in the static test was interpreted to be due to the increase of the factor, k, relating to the stress necessary to unpin the dislocations from their locking atmospheres and due to the increase of the frictional stress, σ_i. These increases are mainly due to a smaller contribution of thermal fluctuation to unpinning of dislocations and dislocation movement in the lattice under dynamic loading owing to the lack of time for the dynamic test. Further, the effect of Mg contents on k was discussed and some dislocation locking mechanisms were considered. The effect of Mg contents on σ_i was also discussed by taking into account the degree of precipitating or solute clustering and dislocation intersection to produce point defects. As a result, clusters or precipitates are pointed out to be an effective factor to determine the value of σ_i. The multiple yielding, as seen in the static test, was not observed under dynamic loading.

View full abstract

The superiority of hot-workability of consumable electrode arc melted metal is examined by the hot torsion test. The test materials are air-melted AISI 316 stainless steel and its consumable electrode arc-remelted steel. The oxygen content decreases remarkably from 0.0084% to 0.0033% and the Mn content decreases a little from 1.84% to 1.4% by arc-remelting, but other contents show no change. The forgings are solution-treated at 1150°C for 1 hr and water-quenched. All the specimens show the microstructures of austenite, and the results of tensile tests and impact tests at room and elevated temperatures show no difference between the air-melted and the arc-melted specimens, except for the reduction in area of the former is slightly higher than that of the latter. Tensile tests for water-quenched specimens from various temperatures of hot torsion tests, also exhibit the same results as above, and their structures are always of austenite, Hot torsion tests are carried out at 650, 800, 900, 1000, 1100 and 1200°C at the rotation speed of 30 rpm and 128 rpm. Revolutions to failure reach a minimum at 800°C and henceforth increase with rise in temperature, while the maximum torque values decrease gradually. In these tests, the arc-melted specimen, which is almost identical with the air-melted one in tensile properties and structures as mentioned above, indicates higher values than the others in the revolutions to failure, the maximum torque value and the area surrounded by the revolutions-torque curve. These results indicate that the consumable-electrode arc-melted material is superior to the air-melted one in hot workability as well as resistance to deformation at elevated temperatures.

View full abstract

In order to illustrate the corrosion behavior of metals under radiation, we undertook a systematic study of the corrosion of pure metals in mineral acid under γ-ray irradiation. In this paper are presented the results of experiments on the corrosion of copper in sulfuric acid. Irradiation cells containing sulfuric acid in the desired concentration in which pure copper plate is immersed is irradiated with a Co-60 γ-ray source, the specimen is hauled up from the sulfuric acid and weighed to determine the weight loss; if necessary, the surface of the specimens are examined photographically, or by the X-ray or electron diffraction method, and the corrosion products are chemically analysed. The test results are compared with those of reference runs conducted under the same conditions minus irradiation. The results are as follows: (1) Effect of radiation is pronounced above the total dose of 10⁷ r. (2) The loss of weight becomes greater with the higher acidity up to 5 N and then decreases, reaching a minimum at 20 N. But in concentrated sulfuric acid, it is remarkable and a dark green corrosion product is precipitated which is assumed to be composed of CuS. (3) The effect of radiation is mainly due to the decomposition of sulfuric acid but not to the influence on the nature of the metal itself. (4) The above results are discussed from the standpoint of the radiolysis of sulfuric acid solution.

View full abstract

A method for producing steel-dies by cold-hubbing has been recently developed. In this method, some important problems remain unsolved on lubricants and tools, because steels have a high resistance to forming. In this report, the results of investigation on the effect of coatings and lubricants are described, and the hubbing depths are also compared and discussed for several steels, using a hardened W-Cr tool steel and a sintered carbide alloy as the hub materials. The lubricating properties are shown by the frictional force per unit contact area in the steady flow stage of hubbing poocess. Manganese phosphate is the most excellent coating on the steel worked. Molybdenum disulphide, stearate of divalent metals and wax such as stearil-alcohol are excellent lubricants when used with a suitable coating. The resistance to hubbing in plain carbon steels increased linearly with the carbon content, but in alloy steels it is higher than in plain carbon steel of the same carbon content. The commercially pure iron made by the L. D. converter method has a very low resistance to hubbing, and this material is to be noticed as an excellent material for cold-forming. Even high-resistance materials to hubbing, such as tool steels in an annealed state, is possible to be hubbed more than 40 mm in depth at a stroke. In such a case, however, the hub material that endures the pressure of not less than 450 kg/mm² should be used.

View full abstract

The density ρ, Young’s modulus E, the coefficient of linear thermal expansion α, and the temperature coefficient of Young’s modulus e were measured with alloys of iron and palladium cooled down in a furnace after heating at 1000°C to ascertain whether a theory of Invar established previously by one of the present investigators is valid in the above alloy system or not. It was found that the both curves of ρ and of E to the alloying concentration showed a slight bend at the composition of about 30 at% of palladium, and also α showed its minimum value of +8.9×10⁻⁶ and e its conspicuous positive maximum of +139.9×10⁻⁵ at 30 at% of palladium. The alloy containing 30 at% of palladium consists almost of the γ phase after furnace cooling, but when it was cooled slower than 30°/hr from 800°C, the γ phase decreased reaching zero at the rate of 5°/hr, accompanying an increase of α to +12.3×10⁻⁶, and a decrease of e to −12.5×10⁻⁵, both in the order of usual alloys. According to these results, it can be confirmed that the theory of Invar mentioned above is also valid in the iron and palladium system under all conditions.

View full abstract