Segmental dynamics of dilute polymer blends

13C and 2H nuclear magnetic resonance has been used to investigate the segmental dynamics of polymers in dilute miscible polymer blends of poly(ethylene oxide) with poly(methyl methacrylate), polyisoprene in series of host polymer matrices, and polystyrene in a series of polymer and small molecule hosts. The segmental dynamics of the dilute polymer was often found to be greater than a factor of 10 faster (or slower) than the host matrix; this dynamic difference was found be as dramatic as 12 orders of magnitude in the poly(ethylene oxide)/poly(methyl methacrylate) blends near the Tg of poly(methyl methacrylate). These results are not intuitively anticipated, but can be accounted for by considering intramolecular effects.; One intramolecular model, the Lodge-McLeish model (LM), was used to describe all dilute blend data. The LM model predicts a single value of self-concentration (&phis;self) that is constant for a given polymer; &phis; self[poly(ethylene oxide)] = 0.15, &phis;self[polyisoprene] = 0.45, and &phis;self[polystyrene] = 0.27. These values of &phis; self in the LM model did not quantitatively predict segmental dynamics in the dilute blends, however the LM model was generally successful in fitting the segmental dynamics of the dilute component; fit values of &phis; self varied dramatically depending on host matrix, ranging from 0.14 to 0.85. The ability of the LM model to fit dilute poly(ethylene oxide), polyisoprene, and polystyrene segmental dynamics argues that &phis;self is independent of temperature, while other work supports the assumption that &phis; self is independent of composition. We varied &phis;self for a given polymer with blending partner in order to obtain quantitative fits to the segmental dynamics of the dilute polymer in the blend. We find that a single value of &phis;self in the framework of the LM model semi-quantitatively predicts the segmental dynamics of the dilute component in miscible polymer blends. We find that self-concentration is the dominant factor that determines polymer segmental dynamics in miscible polymer blends, and the LM model provides a solid foundation from which fully quantitative models are likely to develop.