A critical review of recent experimental studies on elongational flow behavior of polymer melts is presented. Discussion is limited only on the uniaxial elongational flow at a constant rate of strain and at a constant stress. Three types of change in transient elongational viscosity μ(t, ε) with time t have been observed for polystyrene (PS) melts, depending on their molecular weight distribution, where is the rate of strain. The first type is characterized by a smooth and slow growth in μ(t, ε). Polystyrene melts with fairly narrow molecular weight distribution (Mw/Mn=1.2-2.3) show this type of behavior. The second type shows a rapid growth in μ(t, ε) which was obtained for PS melts with broader distribution (Mw/Mn=3.0-4.7). A two-step growth in μ(t, ε) at high rates of strain is characteristic to the third type, which was observed for a bimodal molecular weight distribution. It is shown that two types of elongational flow behavior have been observed in high density polyethylene melts with broad molecular weight distribution. One is characterized by monotonic increase in μ(t, ε) with t and by monotonic decrease in steady elongational viscosity μ(ε) with ε. On the other hand, a two-step growth in μ(t,ε) at high rates of strain and a maximum followed by a remarkable decrease in μ(ε) are observed as the other type. The difference between these two types of behavior seems due to the molecular weight of high density polyethylene. Two types of elongational flow behavior have been found for polypropylene melts depending on the molecular weight distribution. The characteristic behavior of samples with relatively narrow distribution is summarized by (i) monotonic increase in μ(t,ε) with t toward μ(ε), (ii) increase in μ(ε) with a at high ε, and (iii) uniform elongation up to abrupt fracture. The characteristic features of broader distribution polypropylenes are (i) two-step rapid increase in μ(t, ε) with t, (ii) maximum of μ(t, ε) and decrease in the value with increasing ε, and (iii) necking just after the maximum of μ(t, ε) followed by ductile failure. For low density polyethylene melts, effects of branching on elongational flow behavior are not always clear due to the extremely broad molecular weight distribution. There are some measurements which suggest that chain branching contributes to increasing the value of steady elongational viscosity. The elongational flow measurements on model branched polymers with narrow distribution are required to give a quantitative conclusion for this problem. Elongational stress overshoot for a low density polyethylene melt is also discussed.
A steady laminar flow of incompressible Non-Newtonian fluids (power law fluid and Sutterby fluid) in the inlet region of rectangular duct was analyzed by finite difference methods. The main results obtained are as follows. (1) The pressure drop of a Non-Newtonian fluid in the inlet region is smaller than that of a Newtonian fluid and its measure increases with an increase in Non-Newtonian property of the fluid. (2) The velocity in the duct center of a Non-Newtonian fluid in the inlet region is smaller than that of a Newtonian fluid and its measure increases with an increase in Non-Newtonian property of the fluid. (3) The velocity profile in a Non-Newtonian fluid becomes flatter than that in a Newtonian fluid, and the velocity in the inlet region develops retaining flatter profile. (4) The additional pressure loss in the inlet region is smaller in a Non-Newtonian fluid than in a Newtonian fluid and its measure increases with an increase in Non-Newtonian property of the fluid. (5) The inlet length of a Non-Newtonian fluid is larger than that of a Newtonian fluid. (6) The pressure drop and the velocity in the duct center of a Non-Newtonian fluid in the inlet region of a rectangular duct decrease with an increase in the aspect ratio of the duct as those of a Newtonian fluid.
The two-process kinetic theory has been applied for a shear-rate dependence of the steady-shear viscosity of suspensions. The theory proposed by us was successfully used for an interpretation of rheopexy behavior of suspensions. By the use of the two-process model, the contribution of the structural change occuring in suspension has been separated from the total viscosity measured. It has been found that the shear-rate dependence of the steady-shear viscosity consists of two contributions which correspond to the two processes of the structural change. The numerical constants in the equation and the ratios of the constants for the second process to those for first have been evaluated for the suspensions of titanate fibers.
Tensile flow of amorphous poly(vinyl chloride) in the glassy state was examined. The behavior at the lower yield point was compared with the behavior at the upper yield point that had been considered to be a state of pure plastic flow. As was found previously for other polymers, the upper yield point was influenced significantly by the glass transition, while the lower yield point was not. The upper and lower yield points were examined in terms of phenomenologies by Roetling, Bauwens et al. and Brady et al. and of molecular theories by Robertson and by Argon for plastic flow of glassy polymers. No satisfactory agreements of the theories were seen with our results for the upper yield point. However, Robertson's structural temperature theory showed a fairly good agreement with our results obtained at the lower yield point as well as with those for other amorphous polymers reported elsewhere.
Radial distribution of particles in a capillary tube has been studied on neutrally buoyant suspensions of spherical particles flowing through the tube.. The ranges of the solid volume fraction and the relative particle Reynolds number were 0.05 to 0.205 and 0.048 to 2.0, respectively. Two typical particle distribution patterns were observed. One was an accumulation of particles near the center of the tube, and the other was that of the particles near the annular space whose radius was about a half of the tube radius. The former type was observed at low relative particle Reynolds number and at high solid volume fraction. The latter type was observed at high relative particle Reynolds number and at low solid volume fraction. Increase in the relative particle Reynolds number and decrease in the solid volume fraction changed the particle distribution patterns similarly. All patterns of the particle distribution appeared to be a combination of these two types except that in the suspension having solid volume fraction above 0.2.