Capillary penetration of a series of entangled poly(ethylene oxide) melts within nanopores of self-ordered alumina follows an approximate t behavior according to the Lucas-Washburn equation; t is the time. However, the dependence on the capillary diameter deviates from the predicted proportionality to d; d is the pore diameter. We observed a reversal in the dynamics of capillary rise with polymer molecular weight. Chains with 50 entanglements (M ≤ 100 kg/mol) or less show a slower capillary rise than theoretically predicted as opposed to chains with more entanglements (M ≥ 500 kg/mol) that display a faster capillary rise. Although a faster capillary rise has been predicted by theory and observed experimentally, it is the first time to our knowledge that a slower capillary rise is observed for an entangled polymer melt under conditions of strong confinement (with 2R/d = 1). These results are discussed in the light of theoretical predictions for the existence of a critical length scale that depends on the molecular weight and separates the microscopic (d < d) from the macroscopic (d > d) regime.

A unified theory for the imbibition dynamics of entangled polymer melting into nanopores is presented. Experiments demonstrate the validity of t dependence but contradict the predictions of the classical Lucas-Washburn equation because of the prefactor. A reversal in dynamics of capillary filling is reported with increasing polymer molecular weight. Polymer imbibition under nanometer confinement can be discussed by two mechanisms: one is the standard hydrodynamic flow, resulting in a parabolic flow profile. When the inner wall has a strong attraction to the polymer, a layer of immobile chains is created, resulting in an increase of the effective viscosity and to slower imbibition. The other is the reptation model proposed by Johner et al., leading to a plug flow profile and to the reduction in the effective viscosity (faster imbibition). The reversal in dynamics of polymer imbibition can be explained by the competition between these two mechanisms.

No abstract

No abstract

The crystallization and dynamics of water confined in model mesoporous silica particles (pore diameters ranging from 2.1 to 5 nm; pore length ≈ 1 μm) are studied in homogeneous aqueous suspensions by dielectric spectroscopy, differential scanning calorimetry, and nuclear magnetic resonance (NMR) techniques. We establish the phase diagram (T vs 1/d) of confined water covering a broad range of pore diameters. A linear dependence of the heterogeneous and the homogeneous nucleation temperatures on the inverse pore diameter is shown. The two lines converge at a pore diameter of ∼2.6 nm, below which formation of stable crystals is suppressed. By combining dielectric spectroscopy and different NMR techniques, we determine the dynamics of water within mesoporous silica over broad temperature and frequency ranges. Both techniques identify two dynamically distinguishable fractions of confined water coexisting within the pores. We attribute the two fractions to an interfacial water layer at the pore walls and confined water in the pore interior. Two alternative scenarios are proposed to rationalize the coexistence of two dynamically distinguishable water fractions. In the first scenario, two liquid fractions of water coexist under extreme confinement conditions for a range of temperatures; we discuss similarities with the two ultraviscous liquids (high-density liquid and low-density liquid) put forward for supercooled bulk water. In the second scenario, a liquid and a solid phase coexist; we conjecture that highly distorted and unstable crystal nuclei exist under extreme confinement that exhibit reorientation dynamics with time scales intermediate to the surrounding confined liquid and to bulk ice.