Among the interesting problems concerning viscoelasticity of polymer solutions there are three subjects for the present consideration, the hydrodynamic interaction of polymer segments, the non-Newtonian viscosity and entanglement among the polymer chains. (1) Hydrodynamic interaction: As the model of polymer chains the nondraining sphere with the same radius as the radius of gyration of a polymer chain itself seems to be inadequate, seeing that nearly the half number of the segments are located outside the supposed sphere. By adopting the concept of shielding length after the Debye-Bueche's paper, we can derive the relation between the geometrical expansion factor α and the expansion factor for the viscosity αη such as αη3≅α2.43, under some conditions. The precise relation has been derived by Kurata-Yamakawa near Θ-temperature. This indicates that the partially draining model will be better for the model of polymer chains, because the experimental results support this relation. (2) Non-Newtonian viscosity: The experiments show non-Newtonian intrinsic viscosity even at Θ-temperature. Chikahisa applied the equivalent rigid ellipsoid model after Kurata-Stockmayer-Roig's paper to this problem; in this case the non-Newtonian viscosity is due to the orientation of the ellipsoid in a shear field, and may be observed at any temperature. It is pointed out in this review that the conformation change of the chain due to micro-Brownian motion may cause further effect, and it is not easy to show which is dominant for the shear-rate dependence of viscosity. (3) Entanglement; As soon as the concentration of the polymer solution exceeds the order of one percent, the entanglement of polymer chains becomes appreciable. Up to the concentration of the order of ten percents, perhaps a few chains entangle with each other, and make a cluster in the solution. At higher concentrations, the entanglement results in the network formation throughout the system. These situations are supported by the concentration as well as molecular weight dependences of viscosity, equilibrium compliance, dynamic modulus, Weissenberg effect, and so on.
In the thermal fatigue caused by transient temperature gradients in cross-section, the strain range and the degree of thermal strain restraint seem to be affected by the thermal cycling rate. The following study was performed to make clear the effects of heating and cooling rate on the thermal fatigue life and crack growth behavior of 18-8 and 13Cr steels under conditions of transient temperature gradient. The results obtained are summarized as follows: (1) When the heating rate is 100°C/sec (average heating rate to 600°C), the thermal fatigue life N01 decreases from 10% to 70% of the rate at 200°C/sec. This decrease is understood as the result of the growing tensile strain during the cooling process occasioned by the rising internal temperature of the specimen. (2) When the heating rate is 300°C/sec (or 250°C/sec for 13Cr steel), the crack growth ratio rises by the increasing degree of thermal strain restraint, but the thermal fatigue life has about the same value as the rate at 200°C/sec. This is the effect of the tensile stress at elevated temperature. (3) In the case of air cooling, the plastic strain range and the crack growth ratio decrease compared with water cooling, but the thermal fatigue life does not increase perceptively. (4) From these results, it is evident that under conditions of transient temperature gradient the thermal fatigue life is affected not only by the plastic strain range, but also by the working temperature while the strain and thermal stress act on it, and the remarkable thermal fatigue damage is brought about by the tensile stress introduced on the heating surface at the maximum temperature.
As one of the methods of investigation of fatigue strength of metal under service load, the control-apparatus for variable load was added to 1.6t-m rotary bending fatigue testing machine. By means of this machine, the following fatigue tests have been conducted. (1) The relation between α and σw1, σw2 for SS41, large sized V-notched test piece. (a) σw1≅(σw0/α) where, σw0: Fatigue limit of plain test piece. α: Stress concentration factor. σw1: Minimum stress, cracked, at 107 repetitions. (b) σw2≅6kg/mm2 σw2: Maximum stress, without any crack developed, at 107 repetitions. (2) Σ(n/N) of fatigue strength under variable load. (Sine wave and modified rectangular wave) is (0.70∼1.88). where, test pieces are V-notched type (ρ=1mm, d=60mm, D=87mm).
Polymethyl methacrylate specimens were fractured by the static bending, and the speed of the crack propagation was measured by the speed of breaking of the electro-conductive coatings on the surface of the specimens. The rate of crack propagation speed obtained by our method is found to be greater than that which has been obtained by any other investigator. It is found that the fracture morphology is roughly divided into three main classes: (1) the glassy fracture surface containing a number of geometric figures resembling parabolas, (2) the fracture surface having the semicircular“ribs or scallops”of great roughness, and (3) the fracture surface of the greatest roughness at low temperature. It seems that the concentric scallops observed on the fracture surface of the notchless specimens have been produced by the dilatational wave based on the stress release as the crack spreads over the plate.
A new biaxial tension testing machine with flat specimens has been designed and constructed to investigate the plastic behavior of metal under biaxial tension. The capacity of this machine is 5 tons and the stroke is 100mm. The specimen is made in cross-shape. Its dimensions were so determined as to make the region of homogeneous plastic deformation in the middle of the specimen as large as possible. The determination of the stress in the specimen was found possible by the use of the equivalent sectional area. By assuming the yield point as the stress of a permanent set of 200×10-6, an initial and three subsequent yield surfaces were experimentally determined with eleven brass specimens. The following results were obtained from this experiment. The initial yielding follows Mises condition. After prestraining in the principal stress direction: (1) The subsequent yield surfaces are translated in the direction of prestrain. (2) The subsequent yield surfaces do not expand isotropically in any direction. (3) The expansion of subsequent yield surfaces was found quite small.