The predictions of polymer mode coupling theory for the finite size corrections to the transport coefficients of entangled polymeric systems are tested in comparison with various experimental data. It is found that quantitative descriptions of the viscosities, η, dielectric relaxation time, τ , and diffusion coefficients, D, of polymer melts can be achieved with two microscopic structural fit parameters whose values are in the range expected from independent theoretical or experimental information. An explanation for the (apparent) power law behaviors of η, τ , and D in (chemically distinct) melts for intermediate molecular weights as arising from finite size corrections, mainly the self-consistent constraint release mechanism, is given. The variation of tracer dielectric relaxation times from Rouse to reptationlike behavior upon changes of the matrix molecular weight is analyzed. Self-diffusion and tracer diffusion constants of entangled polymer solutions can be explained as well, if one further parameter of the theory is adjusted. The anomalous scaling of the tracer diffusion coefficients in semidilute and concentrated polystyrene solutions, D ∼ N -2.5 , is predicted to arise due to the spatial correlations of the entanglement constraints, termed "constraint porosity". Extensions of the theory to polymer tracer diffusion through poly(vinyl methyl ether) and polyacrylamide gels provide an explanation of the observation of anomalously high molecular weight scaling exponents in a range where the size of the tracer, Rg, already considerably exceeds the gel pore size, g.
IntroductionThe microscopic polymer mode coupling (PMC) theory connects the dynamics of entangled polymeric systems to the underlying equilibrium liquid structure. This description therefore naturally includes corrections arising from the finite molecular weights of tracer or matrix polymers. In the preceding paper, 1 referred to as paper I, the PMC predictions with finite size corrections have been worked out for the transport properties of entangled solutions and melts and for polymer tracer motion in gels. In the present paper these results are compared with data from many experiments. As the PMC approach has been detailed in the preceding paper, 1 it shall be summarized only shortly in this section and then compared to alternative approaches. The resulting formulas of paper I will be referenced by the equation numbers of that paper with a prefix I.The PMC approach of paper I treats the Rouseentangled crossover problem in a simple interpolative manner. The PMC contributions capturing the entanglement corrections are combined with the Rouse, unentangled quantities, see eqs I.52 and I.68 for D, eqs I.55 and I.67 for η, and eqs I.58 and I.70 for τ . Appreciable deviations from the asymptotes for high molecular weights are therefore not explained by corrections of the Rouse dynamics to the asymptotic, entangled motion. In stark contrast to the contour length fluctuation model of Doi, 2,3 the repton model of Rubinstein, 4 and the Rouse-fluc...